WO2015051624A1

WO2015051624A1 - Method and device for rebuilding three-dimensional object and terminal

Info

Publication number: WO2015051624A1
Application number: PCT/CN2014/074590
Authority: WO
Inventors: 刘健庄; 鞠汶奇; 许春景
Original assignee: 华为技术有限公司
Priority date: 2013-10-09
Filing date: 2014-04-02
Publication date: 2015-04-16
Also published as: CN104574515B; US20160217610A1; EP3040944B1; CN104574515A; EP3040944A1; EP3040944A4; US9898860B2

Abstract

Disclosed is a method for rebuilding a three-dimensional object, characterized in that the method comprises: acquiring two-dimensional line drawing information; segmenting, by the two-dimensional line drawing information, the two-dimensional line drawing according to a degree of freedom to obtain at least one line drawing sub-graph, wherein the degree of freedom is the least number of vertices required to be known in order to determine the spatial location of a three-dimensional object which is composed of planes; rebuilding a three-dimensional sub-object according to the line drawing sub-graph; and combining all the three-dimensional sub-objects to obtain the three-dimensional object. The method for rebuilding a three-dimensional object according to the embodiments of the present invention may automatically rebuild a three-dimensional object according to two-dimensional line drawing information. Also disclosed are an apparatus for rebuilding a three-dimensional object and a terminal.

Description

一种三维物体重建的方法、装置和终端 Method, device and terminal for reconstructing three-dimensional object

技术领域本发明涉及图形图像处理技术，特别涉及一种三维物体重建的方法、装置和终端。 TECHNICAL FIELD The present invention relates to graphics image processing techniques, and in particular, to a method, apparatus, and terminal for reconstructing a three-dimensional object.

背景技术三维模型是物体的三维多边形表示，通常用电脑或者其它影片设备进行显示。显示的物体可以是现实世界的实体，也可以是虚构的东西，既可以如原子般大小，也可以拥有很大的尺寸。任何自然界存在的东西都可以用三维模型表 BACKGROUND OF THE INVENTION A three-dimensional model is a three-dimensional polygonal representation of an object, typically displayed using a computer or other film device. The displayed object can be a real-world entity or a fictional object that can be as large as an atom or a large size. Anything that exists in nature can use a three-dimensional model

三维模型经常用专门的三维建模工具软件生成，但是也可以用其它方法生成。作为点和其它信息集合的数据，三维模型可以手工生成，也可以按照一定的算法生成。尽管三维模型通常按照虚拟的方式存在于计算机或者计算机文件中，但是对于普通的用户而言，在纸或屏幕上通过随手勾画二维线画图来表示三维物体是更为贴近普通人表述三维物体的***面地图成为普通人出行必备的向导工具。传统的计算机辅助设计（CAD) ***能够用来创建三维物体。利用 CAD 软件的三维物体建模的方式虽然可以精确绘制三维物体，但是由于它的专业性与繁琐性，需要人工全程参与，工作量非常大，耗时非常多。与之相比，二维线画图简单易画，很久以来一直是人们最习惯的表达三维物体的方式。但是现有的线画图重建三维物体的方法无法解决复杂三维物体重建过程中频繁出现的局部最优的问题，因此这些算法无法应用到重建复杂的三维物体商业应用中。 3D models are often generated using specialized 3D modeling tool software, but can also be generated using other methods. As data for points and other information sets, the 3D model can be generated manually or generated according to a certain algorithm. Although the three-dimensional model usually exists in a virtual way in a computer or computer file, for ordinary users, by drawing a two-dimensional line drawing on paper or on the screen, the three-dimensional object is more close to the ordinary person's expression of the three-dimensional object. Customary, more convenient representation of 3D objects and 3D models. 3D models have an extremely wide range of applications. In fact, the application of real 3D models is earlier than the application of personal computers. The popularity of personal computers has made the application of 3D models reproduced and controlled by computers popular among all walks of life and thousands of households. Now, 3D models have been used in a variety of different fields. Use them in the medical industry to make accurate models of organs; the film industry uses them for active characters, objects, and real-life movies; the video game industry uses them as resources in computer and video games; as an accurate model of compounds in science The construction industry uses them to display proposed buildings or landscapes; the engineering community uses them in the design of new equipment, vehicles, structures, etc.; in recent decades, the earth sciences have also begun to build 3D geological models; The future 3D map will also replace the popular flat maps. A must-have guide tool for people to travel. Traditional computer-aided design (CAD) systems can be used to create three-dimensional objects. Although the 3D object modeling method using CAD software can accurately draw 3D objects, due to its professionalism and cumbersomeness, it requires manual participation, and the workload is very large and time consuming. In contrast, 2D line drawing is simple and easy to draw, and has long been the most popular way to express 3D objects. However, the existing method of reconstructing three-dimensional objects by line drawing cannot solve the problem of local optimal occurrence frequently in the process of complex three-dimensional object reconstruction. Therefore, these algorithms cannot be applied to reconstruct complex three-dimensional object commercial applications.

发明内容有鉴于此，为解决上述问题，本发明的实施例提供了一种三维物体重建的方法，从而实现了对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便的效果。本发明第一方面的实施例公开了一种三维物体重建的方法，所述方法包括：获取三维物体的二维线画图信息，所述二维线画图信息包括二维线画图；根据所述二维线画图信息，按照自由度分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数；根据所述线画图子图重建三维子物体；组合全部的所述三维子物体得到三维物体。根据本发明实施例的三维物体重建的方法，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。在本发明第一方面的第一实施例中，每一个线画图子图对应一个三维子物体；在全部所述线画图子图中，所述线画图子图所表示的所述三维子物体的最小的自由度为 4。结合本发明第一方面实施例或第一方面的第一实施例的本发明第一方面的第二实施例中，所述根据所述二维线画图信息，按照自由度分割所述二维线画图得到至少一个线画图子图包括：根据所述二维线画图信息得到二维线画图的顶点坐标、顶点之间的线段信息，根据所述二维线画图的顶点坐标和所述顶点之间的线段信息得到二维线画图的面信息；根据所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图。结合本发明第一方面第二实施例的本发明第一方面的第三实施例中，所述二维线画图的面信息包括二维线画图的面；所述根据所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图，具体包括：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。结合本发明第一方面第三实施例的本发明第一方面的第四实施例中，如果所述第一集合中的确定面的数目不小于三，则所述方法还包括：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。结合上述任意之一实施例的本发明第一方面的第五实施例中，所述所述根据所述线画图子图重建三维子物体包括：根据最小化目标函数得到所述线画图子图的顶点的深度值；获取所述线画图子图的顶点坐标；获取三维子物体的面； SUMMARY OF THE INVENTION In view of the above, in order to solve the above problems, an embodiment of the present invention provides a method for reconstructing a three-dimensional object, thereby realizing reasonable segmentation and reconstruction of a line drawing of an object composed of planes, and two-dimensional line drawing. The input method is relatively simple. An embodiment of the first aspect of the present invention discloses a method for reconstructing a three-dimensional object, the method comprising: acquiring two-dimensional line drawing information of a three-dimensional object, wherein the two-dimensional line drawing information includes a two-dimensional line drawing; Dimension drawing information, dividing the two-dimensional line drawing according to degrees of freedom to obtain at least one line drawing sub-picture, wherein the degree of freedom is a minimum vertice required to determine a spatial position of a three-dimensional object composed of planes Reconstructing a three-dimensional sub-object according to the line drawing sub-picture; combining all of the three-dimensional sub-objects to obtain a three-dimensional object. According to the method for reconstructing a three-dimensional object according to an embodiment of the present invention, a line drawing of an object composed of planes can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. In a first embodiment of the first aspect of the present invention, each line drawing subpicture corresponds to a three-dimensional sub-object; in all of the line drawing sub-pictures, the three-dimensional sub-object represented by the line drawing sub-picture Most The small degree of freedom is 4. With reference to the first embodiment of the first aspect of the present invention or the first embodiment of the first aspect of the first aspect of the first aspect, the dividing the two-dimensional line according to the degree of freedom according to the two-dimensional line drawing information Drawing at least one line drawing subgraph includes: obtaining vertex coordinates of the two-dimensional line drawing and line segment information between the vertices according to the two-dimensional line drawing information, according to the vertex coordinates of the two-dimensional line drawing and the vertices The line segment information obtains the surface information of the two-dimensional line drawing; and according to the surface information of the two-dimensional line drawing, the two-dimensional line drawing is divided according to the degree of freedom to obtain at least one of the line drawing sub-pictures. In a third embodiment of the first aspect of the present invention, in combination with the second embodiment of the first aspect of the present invention, the face information of the two-dimensional line drawing includes a face of a two-dimensional line drawing; the drawing according to the two-dimensional line drawing The surface information is divided into the two-dimensional line drawing according to the degree of freedom, and at least one of the line drawing sub-pictures is obtained, which specifically includes: selecting any pair of the two-dimensional line drawing faces that have not been selected yet Adjacent faces, as a first line drawing surface and a second line drawing surface; setting the first set to an empty set, calibrating the first line drawing surface and the second line drawing surface to a determining surface and adding to the And arranging, in all of the faces of the two-dimensional line drawing, other line drawing surfaces except the first line drawing surface and the second line drawing surface as an indeterminate surface; Selecting a third line drawing surface, calibrating the third line drawing surface as a determining surface and adding to the first set, repeating the step until traversing the incomplete surface of the complete portion, wherein the third line drawing surface At least three are not collinear a vertex, and the vertices of the non-collinear line are vertices of the line drawing surface in the first set; obtaining a corresponding line drawing sub-picture according to the determining surface in the first set; repeating the above steps until the All adjacent faces in the face of all 2D line drawings are selected I chose it once. In a fourth embodiment of the first aspect of the present invention, in combination with the third embodiment of the first aspect of the present invention, if the number of the determining faces in the first set is not less than three, the method further includes: a line drawing subgraph represented by a set is added to the second set; a coexistence relationship graph is obtained according to all the line drawing subgraphs in the second set, wherein a vertex of the coexistence relationship graph represents one of the line drawing a sub-picture, the number of faces of the line drawing sub-picture is used as a weight of the vertex, and if there is a coexistence relationship between the two line drawing sub-pictures, corresponding to the two line drawing sub-pictures There is a connection between the vertices; the coexistence relationship diagram represents a coexistence relationship between objects represented by the line drawing subgraph, and the coexistence relationship refers to an object represented by the two line drawing subgraphs The common surface between the objects represented by the two line drawing subgraphs only exists on the same plane; the maximum weight group is found in the coexistence relationship diagram, and the maximum weight is determined. Value group a line drawing subgraph corresponding to the vertex, wherein the group is a picture having vertices and edges, and any two vertices in the group have an edge connection, and the maximum weight group is in the coexistence relationship The group with the largest sum of weights in all the groups of the graph. In a fifth embodiment of the first aspect of the present invention, in combination with the line drawing subgraph, the reconstructing the three-dimensional sub-object includes: obtaining the line drawing sub-picture according to the minimized objective function a depth value of the vertex; acquiring a vertex coordinate of the line drawing subgraph; acquiring a surface of the three-dimensional sub-object;

根据所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。 The three-dimensional sub-object is obtained according to a depth value of a vertex of the line drawing sub-picture, a vertex coordinate of the line drawing sub-picture, and a face of the three-dimensional sub-object.

结合上述任意之一实施例的本发明第一方面的第六实施例中，所述组合全部的所述三维子物体得到所述三维物体，包括：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。结合本发明第一方面第六实施例的本发明第一方面的第七实施例中，所述通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，具体包括：根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。结合上述任意之一实施例的本发明第一方面的第八实施例中，通过下列之一的方式获取所述三维物体的二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式和通过从预先存储的照片中提取线画图的方式。本发明第二方面的实施例公开了一种三维物体重建的方法，所述方法包括：获取三维物体的二维线画图，以及所述二维线画图中的顶点坐标、顶点之间的线段信息；根据所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图；根据每一个所述线画图子图重建一个对应的三维子物体；组合全部的所述三维子物体得到三维物体。根据本发明实施例的三维物体重建的方法，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。在本发明第二方面的第一实施例中，所述根据所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图，具体包括：根据所述二维线画图的顶点和所述顶点之间的线段信息获得二维线画图的面信息；根据所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图。结合本发明第二方面的第一实施例的本发明第二方面的第二实施例中，所述根据所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图包括：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。结合本发明第二方面第二实施例的本发明第二方面的第三实施例中，如果所述第一集合中的确定面的数目不小于三，则在根据所述第一集合中的确定面得到一个对应的线画图子图之后，所述方法还包括：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。结合上述本发明第二方面任意之一实施例的本发明第二方面的第四实施例中，所述根据一个所述线画图子图重建一个对应的三维子物体，包括：根据最小化目标函数得到所述线画图子图的顶点的深度值；获取所述线画图子图的顶点坐标；获取三维子物体的面；根据所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。结合上述本发明第二方面任意之一实施例的本发明第二方面的第五实施例中，所述组合全部的所述三维子物体得到所述三维物体，包括：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。结合本发明第二方面第五实施例的本发明第二方面的第六实施例中，所述通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，具体包括：根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。结合上述本发明第二方面任意之一实施例的本发明第二方面的第七实施例中，通过下列之一的方式获取所述二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式、通过从预先存储的照片中提取线画图的方式。本发明第三方面的实施例公开了一种三维物体重建的装置，所述装置包括：获取模块，用于获取三维物体的二维线画图信息，所述二维线画图信息包括二维线画图；分割模块，用于根据所述获取模块获取的所述二维线画图信息，按照自由度分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数；重建模块，用于根据所述分割模块得到的所述线画图子图重建三维子物体；组合模块，用于组合全部的所述三维子物体重建模块得到的所述三维子物体得到三维物体。根据本发明实施例的三维物体重建的装置，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。在本发明第三方面的第一实施例中，所述分割模块包括：面信息获取单元，用于根据所述获取模块获得的所述二维线画图信息得到二维线画图的顶点坐标、顶点之间的线段信息，根据所述二维线画图的顶点坐标和所述顶点之间的线段信息得到二维线画图的面信息；面分割单元，用于根据所述面信息获取单元得到的所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图。结合本发明第三方面的第一实施例的本发明第三方面的第二实施例中，所述面分割单元具体用于：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。结合本发明第三方面的第二实施例的本发明第三方面的第三实施例中，如果所述第一集合中的确定面的数目不小于三，则所述面分割单元还用于：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。 In a sixth embodiment of the first aspect of the present invention, in combination with any one of the above embodiments, the combining the all three-dimensional sub-objects to obtain the three-dimensional object comprises: combining all the three-dimensional objects by means of site bonding The object, the three-dimensional object is obtained, and the manner of combining the parts includes at least one of the following: point combination, point line combination, point surface combination, line bonding, line surface bonding, face bonding Combined with the dotted line. In a seventh embodiment of the first aspect of the present invention, in combination with the sixth embodiment of the first aspect of the present invention, the three-dimensional object is combined by means of a combination of parts to obtain the three-dimensional object, and specifically includes: The two-dimensional line drawing information is used to obtain the combined part information of the three-dimensional sub-object; the size of the three-dimensional sub-object is adjusted according to the combined part information of the three-dimensional sub-object, and the adjusted three-dimensional sub-object is obtained; The three-dimensional sub-object obtains an initial three-dimensional object; the initial three-dimensional object is adjusted according to a minimized objective function to obtain the three-dimensional object. With reference to the eighth embodiment of the first aspect of the present invention in any one of the above embodiments, the two-dimensional line drawing information of the three-dimensional object is acquired by one of the following methods: a method of capturing by a camera device, and input by a touch device The mode, the way the mouse is input, and the way the line drawing is extracted from pre-stored photos. An embodiment of the second aspect of the present invention discloses a method for reconstructing a three-dimensional object, the method comprising: acquiring a two-dimensional line drawing of a three-dimensional object, and coordinates of a vertex in the two-dimensional line drawing and line segment information between the vertices Dividing the two-dimensional line drawing according to the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices to obtain at least one line drawing sub-picture; reconstructing a corresponding one according to each of the line drawing sub-pictures a three-dimensional sub-object; combining all of the three-dimensional sub-objects to obtain a three-dimensional object. According to the method for reconstructing a three-dimensional object according to an embodiment of the present invention, a line drawing of an object composed of planes can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. In a first embodiment of the second aspect of the present invention, the dividing the two-dimensional line drawing according to the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices to obtain at least one line drawing sub-picture Specifically, including: Obtaining surface information of the two-dimensional line drawing according to the line segment information between the vertices of the two-dimensional line drawing and the vertices; and dividing the two-dimensional line drawing according to the surface information of the two-dimensional line drawing to obtain at least one The line drawing subgraph. In a second embodiment of the second aspect of the present invention, in combination with the first embodiment of the second aspect of the present invention, the two-dimensional line drawing is segmented according to the surface information of the two-dimensional line drawing to obtain at least one of the The line drawing sub-picture includes: selecting any pair of adjacent faces that have not been selected in the faces of all the two-dimensional line drawing, as the first line drawing surface and the second line drawing surface; setting the first set to be empty a set, the first line drawing surface and the second line drawing surface are marked as a determining surface and added to the first set; the first line is removed in all the faces of the two-dimensional line drawing The drawing surface and the other line drawing surface of the second line drawing surface are marked as an indeterminate surface; the third line drawing surface is selected in the indeterminate surface, and the third line drawing surface is marked as a determining surface and added to the first In a set, the steps are repeated until the indeterminate surface of the complete portion is traversed, wherein the third line drawing surface has at least three vertices that are not collinear, and the vertices of the non-collinear lines are the first set The vertices of the line drawing surface; The determining face in the first set obtains a corresponding line drawing sub-picture; repeating the above steps until all adjacent faces in the faces of the all two-dimensional line drawing are selected once. In a third embodiment of the second aspect of the present invention in combination with the second embodiment of the second aspect of the present invention, if the number of the determination faces in the first set is not less than three, the determination in the first set is After obtaining a corresponding line drawing sub-picture, the method further comprises: adding the line drawing sub-picture represented by the first set to the second set; according to all the line drawing sub-pictures in the second set Obtaining a coexistence relationship graph, wherein the coexistence One vertex of the relationship graph represents one of the line drawing subgraphs, and the number of faces of the line drawing subgraph is used as the weight of the vertex, if there is a coexistence relationship between the two line drawing subgraphs Correspondingly, there is a connection between the vertices corresponding to the two line drawing sub-pictures; the coexistence relationship diagram represents a coexistence relationship between the objects represented by the line drawing sub-picture, and the coexistence relationship refers to the The planes represented by the two line drawing subgraphs do not share any plane or the common plane between the objects represented by the two line drawing subgraphs only exist on the same plane; in the coexistence relationship diagram Finding a maximum weight group, determining a line drawing subgraph corresponding to a vertex of the maximum weight group, wherein the group is a graph having vertices and edges, and any two vertices in the group have an edge connection The maximum weight group is the group with the largest sum of weights in all groups of the coexistence relationship graph. In a fourth embodiment of the second aspect of the present invention, in combination with any one of the foregoing second embodiments of the present invention, the reconstructing a corresponding three-dimensional sub-object according to one of the line drawing sub-pictures comprises: according to a minimum objective function Obtaining a depth value of a vertex of the line drawing sub-picture; acquiring vertex coordinates of the line drawing sub-picture; acquiring a surface of the three-dimensional sub-object; and determining a depth value of the vertex of the line drawing sub-picture, the line drawing sub-picture The vertex coordinates and the faces of the three-dimensional sub-objects result in the three-dimensional sub-object. In combination with the fifth embodiment of the second aspect of the present invention in any one of the foregoing second aspects of the present invention, the combining the all three-dimensional sub-objects to obtain the three-dimensional object includes: combining all by means of part combination The three-dimensional sub-objects obtain the three-dimensional object, and the manner in which the parts are combined includes at least one of the following: point combination, point line bonding, point surface bonding, line bonding, line surface bonding, face bonding, and Point line surface combination. In a sixth embodiment of the second aspect of the present invention, in combination with the fifth embodiment of the second aspect of the present invention, the three-dimensional object is combined by means of a combination of parts to obtain the three-dimensional object, and the specific package is obtained. And: obtaining the combined part information of the three-dimensional sub-object according to the two-dimensional line drawing information; adjusting the size of the three-dimensional sub-object according to the combined part information of the three-dimensional sub-object, and obtaining the adjusted three-dimensional sub-object; The adjusted three-dimensional sub-object obtains an initial three-dimensional object; and the initial three-dimensional object is adjusted according to a minimized objective function to obtain the three-dimensional object. In a seventh embodiment of the second aspect of the present invention, in combination with any one of the foregoing second embodiments of the present invention, the two-dimensional line drawing information is obtained by one of the following methods: by means of a camera device, by a touch device The method of input, the way of input by mouse, and the way of drawing a line drawing from a pre-stored photo. An embodiment of the third aspect of the present invention discloses a device for reconstructing a three-dimensional object, the device comprising: an acquiring module, configured to acquire two-dimensional line drawing information of a three-dimensional object, wherein the two-dimensional line drawing information includes a two-dimensional line drawing a segmentation module, configured to divide the two-dimensional line drawing according to degrees of freedom according to the two-dimensional line drawing information acquired by the acquiring module, to obtain at least one line drawing sub-picture, wherein the degree of freedom is determined by one a minimum number of vertices that need to be known for the spatial position of the three-dimensional object composed of the plane; a reconstruction module for reconstructing the three-dimensional sub-object according to the line drawing sub-picture obtained by the segmentation module; a combination module, for combining all The three-dimensional sub-object obtained by the three-dimensional sub-object reconstruction module obtains a three-dimensional object. According to the apparatus for reconstructing a three-dimensional object according to the embodiment of the present invention, the line drawing of the object composed of the plane can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. In a first embodiment of the third aspect of the present invention, the segmentation module includes: a surface information acquiring unit, configured to obtain vertex coordinates of the two-dimensional line drawing and line segment information between the vertices according to the two-dimensional line drawing information obtained by the acquiring module, according to the vertex coordinates of the two-dimensional line drawing and the The line segment information between the vertices obtains the surface information of the two-dimensional line drawing; the surface segmentation unit is configured to segment the two-dimensional line drawing according to the face information, and divide the two-dimensional according to the degree of freedom Line drawing, at least one of the line drawing subgraphs is obtained. In a second embodiment of the third aspect of the present invention, in combination with the first embodiment of the third aspect of the present invention, the surface segmentation unit is specifically configured to: select any one of the faces of the two-dimensional line drawing An adjacent surface that has not been selected, as a first line drawing surface and a second line drawing surface; setting the first set to an empty set, and calibrating the first line drawing surface and the second line drawing surface to a certain plane and adding Into the first set; calibrating other line drawing surfaces of the first line drawing surface and the second line drawing surface as indeterminate faces in all faces of the two-dimensional line drawing; Selecting a third line drawing surface in the indefinite surface, calibrating the third line drawing surface as a determining surface and adding to the first set, repeating the step until the indeterminate surface of the complete portion is traversed, wherein The third line drawing surface has at least three vertices that are not collinear, and the vertices of the non-collinear lines are vertices of the line drawing surface in the first set; and a corresponding line is obtained according to the determining surface in the first set Drawing subgraph; repeating the above , Until all the adjacent face of the whole two-dimensional plane of the drawing lines are selected once. In a third embodiment of the third aspect of the present invention, in combination with the second embodiment of the third aspect of the present invention, if the number of the determined faces in the first set is not less than three, the face segmentation unit is further configured to: Adding a line drawing subgraph represented by the first set to the second set; obtaining a coexistence relationship graph according to all the line drawing subgraphs in the second set, wherein one vertex of the coexistence relationship graph represents one The line drawing sub-picture, the face of the line drawing sub-picture The number is used as the weight of the vertex. If there is a coexistence relationship between the two line drawing subgraphs, there is a connection between the vertices corresponding to the two line drawing subgraphs; the coexistence relationship diagram represents a coexistence relationship between the objects represented by the line drawing subgraph, wherein the coexistence relationship means that no one plane or the two line drawing subgraphs are shared between the objects represented by the two line drawing subgraphs. The common surface between the represented objects exists only on the same plane; the maximum weight group is found in the coexistence relationship graph, and the line drawing subgraph corresponding to the vertex of the maximum weight group is determined, wherein A graph having vertices and edges, wherein any two vertices in the cluster have an edge connection, and the maximum weight group is the group with the largest sum of weights in all groups of the coexistence relationship graph.

结合上述本发明第三方面任意之一实施例的本发明第三方面的第四实施例中，所述重建模块包括：重建信息获取单元，用于根据最小化目标函数得到所述线画图子图的顶点的深度值，获取所述线画图子图的顶点坐标，获取三维子物体的面；重建单元，用于根据所述重建信息获取单元得到的所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子 In the fourth embodiment of the third aspect of the present invention, in combination with any one of the foregoing third embodiments of the present invention, the reconstruction module includes: a reconstruction information acquiring unit, configured to obtain the line drawing sub-picture according to the minimum objective function a depth value of the vertex of the line drawing, obtaining a vertex coordinate of the line drawing subgraph, acquiring a surface of the three-dimensional sub-object; and a reconstruction unit, configured to obtain, according to the depth value of the vertex of the line drawing sub-picture obtained by the reconstruction information acquiring unit, The vertex coordinates of the line drawing subgraph and the face of the three-dimensional sub-object obtain the three-dimensional sub-object

结合上述本发明第三方面任意之一实施例的本发明第三方面的第五实施例中，所述组合模块具体用于：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。 In combination with the fifth embodiment of the third aspect of the present invention, the combination module is specifically configured to: combine all of the three-dimensional sub-objects by means of a part combination to obtain the The three-dimensional object, the manner in which the parts are combined includes at least one of the following: point combination, point line bonding, point surface bonding, wire bonding, line bonding, face bonding, and dot line bonding.

结合上述本发明第三方面任意之一实施例的本发明第三方面的第六实施例中，所述组合模块具体用于： In a sixth embodiment of the third aspect of the present invention, in combination with any one of the foregoing third embodiments of the present invention, the combination module is specifically configured to:

根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。 Obtaining joint location information of the three-dimensional sub-object according to the two-dimensional line drawing information; Adjusting the size of the three-dimensional sub-object according to the joint position information of the three-dimensional sub-object to obtain an adjusted three-dimensional sub-object; combining the adjusted three-dimensional sub-object to obtain an initial three-dimensional object; and adjusting the initial according to a minimum objective function A three-dimensional object obtains the three-dimensional object.

结合上述本发明第三方面、第三方面第一至第五任意之一实施例的本发明第三方面的第七实施例中，所述获取模块通过下列之一的方式获取所述三维物体的二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式和通过从预先存储的照片中提取线画图的方式。本发明第四方面的实施例公开了一种三维物体重建的装置，所述装置包括： With reference to the seventh embodiment of the third aspect of the present invention, in combination with the third aspect of the present invention, the acquiring module, the acquiring module acquires the three-dimensional object by one of the following methods. Two-dimensional line drawing information: a method of shooting by a camera device, a method of input by a touch device, a method of input by a mouse, and a method of extracting a line drawing from a pre-stored photo. An embodiment of the fourth aspect of the invention discloses a device for reconstructing a three-dimensional object, the device comprising:

信息获取模块，用于获取三维物体的二维线画图，以及所述二维线画图中的顶点坐标、顶点之间的线段信息；线画图子图分割模块，用于根据所述信息获取模块获取的所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图；三维子物体重建模块，用于根据所述线画图子图分割模块得到的每一个所述线画图子图重建一个对应的三维子物体；三维物体组合模块，用于组合所述三维子物体重建模块重建的全部的所述三维子物体得到三维物体。根据本发明实施例的三维物体重建的装置，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。在本发明第四方面的第一实施例中，所述线画图子图分割模块，包括：二维线画图的面信息获取单元，根据所述信息获取模块获取的二维线画图的顶点和所述顶点之间的线段信息获得二维线画图的面信息；线画图子图分割单元，用于根据所述二维线画图的面信息获取单元获取的所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图。结合本发明第四方面的第一实施例的本发明第四方面的第二实施例中，所述线画图子图分割单元具体用于：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。结合本发明第四方面的第二实施例的本发明第四方面的第三实施例中，所述线画图子图分割单元还用于：如果所述第一集合中的确定面的数目不小于三，则将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。结合上述本发明第四方面任意之一实施例的本发明第四方面的第四实施例中，所述三维子物体重建模块包括：三维子物体重建信息获取单元，用于根据最小化目标函数得到所述线画图子图的顶点的深度值，获取所述线画图子图的顶点坐标，获取三维子物体的面；三维子物体重建单元，用于根据所述三维子物体重建信息获取单元得到的所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。结合上述本发明第四方面任意之一实施例的本发明第四方面的第五实施例中，所述三维物体组合模块具体用于：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式： An information acquiring module, configured to acquire a two-dimensional line drawing of the three-dimensional object, and the vertex coordinates and the line segment information between the vertices in the two-dimensional line drawing; the line drawing sub-graph dividing module is configured to acquire according to the information acquiring module Dividing the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices to segment the two-dimensional line drawing to obtain at least one line drawing sub-picture; a three-dimensional sub-object reconstruction module, configured to draw a picture according to the line drawing Each of the line drawing subgraphs obtained by the graph segmentation module reconstructs a corresponding three-dimensional sub-object; and the three-dimensional object combining module is configured to combine all the three-dimensional sub-objects reconstructed by the three-dimensional sub-object reconstruction module to obtain a three-dimensional object. According to the apparatus for reconstructing a three-dimensional object according to the embodiment of the present invention, the line drawing of the object composed of the plane can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. In a first embodiment of the fourth aspect of the present invention, the line drawing sub-picture segmentation module includes: a surface information acquisition unit of a two-dimensional line drawing, and a vertex and a location of the two-dimensional line drawing acquired according to the information acquisition module Obtaining the surface information of the two-dimensional line drawing by the line segment information between the vertices; a line drawing sub-picture dividing unit, configured to divide the two-dimensional line drawing according to the surface information of the two-dimensional line drawing acquired by the surface information acquiring unit of the two-dimensional line drawing, to obtain at least one of the line drawing sub-pictures . In a second embodiment of the fourth aspect of the present invention, in combination with the first embodiment of the fourth aspect of the present invention, the line drawing sub-picture dividing unit is specifically configured to: select any of the faces of the two-dimensional line drawing a pair of adjacent faces that have not been selected, as a first line drawing surface and a second line drawing surface; setting the first set to an empty set, and calibrating the first line drawing surface and the second line drawing surface to determine And adding to the first set; arranging other line drawing surfaces except the first line drawing surface and the second line drawing surface as an indeterminate surface in all faces of the two-dimensional line drawing; Selecting a third line drawing surface in the indeterminate surface, calibrating the third line drawing surface as a determining surface and adding to the first set, repeating the step until traversing the indeterminate surface of the complete portion, wherein The third line drawing surface has at least three vertices that are not collinear, and the vertices of the non-collinear line are vertices of the line drawing surface in the first set; and one is obtained according to the determining surface in the first set Corresponding line drawing subgraph; heavy Ho above step until all the two-dimensional line drawing of all adjacent facing surfaces are selected once. In a third embodiment of the fourth aspect of the present invention, in combination with the second embodiment of the fourth aspect of the present invention, the line drawing sub-picture dividing unit is further configured to: if the number of determining planes in the first set is not less than Third, adding the line drawing sub-picture represented by the first set to the second set; obtaining a coexistence relationship diagram according to all the line drawing sub-pictures in the second set, wherein one of the coexistence relationship diagrams The vertex represents one of the line drawing sub-pictures, and the number of faces of the line drawing sub-picture is used as the weight of the vertex, and if there is a coexistence relationship between the two line drawing sub-pictures, the corresponding There is a connection between the vertices of the two line drawing sub-pictures; the coexistence relationship diagram represents a coexistence relationship between objects represented by the line drawing sub-picture, and the coexistence relationship refers to the two lines The plane represented by the drawing subgraph does not share any plane or the common plane between the objects represented by the two line drawing subgraphs exists only on the same plane; find the maximum weight in the coexistence relationship diagram a value group, determining a line drawing subgraph corresponding to a vertex of the maximum weight group, wherein the group is a picture having vertices and edges, and an edge connection is formed between any two vertices in the group The maximum weight group is the group with the largest sum of weights in all the groups of the coexistence relationship graph. In a fourth embodiment of the fourth aspect of the present invention, in combination with the fourth aspect of the present invention, the three-dimensional sub-object reconstruction module includes: a three-dimensional sub-object reconstruction information acquisition unit, configured to obtain a minimum objective function a depth value of a vertex of the line drawing subgraph, acquiring a vertex coordinate of the line drawing subgraph, and acquiring a surface of the three-dimensional sub-object; a three-dimensional sub-object reconstruction unit, configured to obtain the unit according to the three-dimensional sub-object reconstruction information acquiring unit The depth value of the vertices of the line drawing subgraph, the vertex coordinates of the line drawing subgraph, and the face of the three-dimensional sub-object obtain the three-dimensional sub-object. In a fifth embodiment of the fourth aspect of the present invention, in combination with any one of the fourth aspects of the present invention, the three-dimensional object combination module is specifically configured to: combine all of the three-dimensional sub-objects by means of part combination, and obtain The three-dimensional object, the manner in which the parts are combined includes at least one of the following:

点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。结合上述本发明第四方面、第四方面第一至第四任意之一实施例的本发明第三方面的第六实施例中，所述三维物体组合模块具体用于：根据所述二维线画图信息得到所述三维子物体的结合部位信息； Dot combination, point and line bonding, point and face bonding, wire bonding, line bonding, face bonding, and point and line bonding. In combination with the sixth embodiment of the third aspect of the present invention, the fourth aspect of the present invention, the three-dimensional object combination module is specifically configured to: according to the two-dimensional line Drawing information to obtain joint position information of the three-dimensional sub-object;

根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。本发明第五方面的实施例公开了一种三维物体重建的终端，所述终端包括：如本发明第三方面实施例和第四方面实施例任意之一所述的三维物体重建装置，显示器，其中，所述显示器用于显示所述三维物体。 Adjusting the size of the three-dimensional sub-object according to the joint location information of the three-dimensional sub-object, and obtaining a tone The entire three-dimensional sub-object is combined; the adjusted three-dimensional sub-object is combined to obtain an initial three-dimensional object; and the initial three-dimensional object is adjusted according to a minimized objective function to obtain the three-dimensional object. The embodiment of the fifth aspect of the present invention discloses a terminal for reconstructing a three-dimensional object, the terminal comprising: a three-dimensional object reconstruction device according to any one of the third aspect of the present invention and the fourth aspect, a display, Wherein the display is used to display the three-dimensional object.

附图说明为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其他的附图。图 1为根据本发明实施例的一种三维物体重建方法的流程图。图 2为根据本发明另一实施例的一种三维物体重建方法的流程图。图 3为根据本发明实施例的一种三维物体重建的具体示例。图 4为根据本发明实施例的另一种三维物体重建的具体示例。图 5为根据本发明实施例的将基本物体组合为三维物体的具体示例。图 6为根据本发明实施例的通过线段将基本物体组合为三维物体的具体示例。图 7为根据本发明实施例的通过照片获取二维线画图信息重建三维物体的具体示例。图 8为根据本发明实施例重建的三维物体的不同角度的效果示意图。图 9为根据本发明实施例计算方式得到线画图的面的一种示例。图 10为根据本发明实施例的一种三维物体重建方法的流程图。图 11为根据本发明实施例的一种维物体重建装置的示意图。图 12为根据本发明另一实施例的种三维物体重建装置的示意图。图 13为根据本发明实施例的一种维物体重建终端的示意图。 BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the embodiments or the description of the prior art will be briefly described below, and obviously, in the following description The drawings are only some of the embodiments of the present invention, and other drawings may be obtained from those skilled in the art without departing from the drawings. 1 is a flow chart of a method for reconstructing a three-dimensional object according to an embodiment of the present invention. 2 is a flow chart of a method for reconstructing a three-dimensional object according to another embodiment of the present invention. FIG. 3 is a specific example of three-dimensional object reconstruction according to an embodiment of the present invention. 4 is a specific example of another three-dimensional object reconstruction in accordance with an embodiment of the present invention. FIG. 5 is a specific example of combining basic objects into three-dimensional objects according to an embodiment of the present invention. 6 is a specific example of combining basic objects into three-dimensional objects by line segments, in accordance with an embodiment of the present invention. FIG. 7 is a specific example of reconstructing a three-dimensional object by acquiring two-dimensional line drawing information through a photo according to an embodiment of the present invention. FIG. 8 is a schematic diagram showing the effect of different angles of a three-dimensional object reconstructed according to an embodiment of the present invention. FIG. 9 is an illustration of a face of a line drawing obtained by a calculation method according to an embodiment of the present invention. FIG. 10 is a flow chart of a method for reconstructing a three-dimensional object according to an embodiment of the present invention. 11 is a schematic diagram of a dimensional object reconstruction device in accordance with an embodiment of the present invention. FIG. 12 is a schematic diagram of a three-dimensional object reconstruction apparatus according to another embodiment of the present invention. FIG. 13 is a schematic diagram of a dimensional object reconstruction terminal according to an embodiment of the invention.

具体实施方式下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。 The technical solutions in the embodiments of the present invention are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present invention. It is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. example. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

下面结合图 1描述根据本发明实施例的一种三维物体重建的方法，如图 1 所示，所述方法包括： A method for reconstructing a three-dimensional object according to an embodiment of the present invention is described below with reference to FIG. 1. As shown in FIG. 1, the method includes:

S11 : 获取三维物体的二维线画图信息，所述二维线画图信息包括二维线画图。 S11: Obtain two-dimensional line drawing information of the three-dimensional object, and the two-dimensional line drawing information includes a two-dimensional line drawing.

S12: 根据自由度和所述二维线画图信息分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数。 S12: segmenting the two-dimensional line drawing according to the degree of freedom and the two-dimensional line drawing information to obtain at least one line drawing sub-picture, wherein the degree of freedom needs to be known to determine a spatial position of a three-dimensional object composed of a plane. The minimum number of vertices.

S13 : 根据所述线画图子图重建三维子物体。 S13: reconstruct a three-dimensional sub-object according to the line drawing sub-picture.

S14: 组合全部的所述三维子物体得到所述三维物体。根据本发明实施例的三维物体的重建方法，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。下面结合图 10描述根据本发明另一实施例的一种三维物体重建的方法，如图 10所示， S14: Combine all the three-dimensional sub-objects to obtain the three-dimensional object. According to the reconstruction method of the three-dimensional object according to the embodiment of the present invention, the line drawing of the object composed of the plane can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. A method for reconstructing a three-dimensional object according to another embodiment of the present invention will be described below with reference to FIG. 10, as shown in FIG.

S101 : 获取三维物体的二维线画图，以及所述二维线画图中的顶点坐标、顶点之间的线段信息。 S102: 根据所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图。 S101: Obtain a two-dimensional line drawing of the three-dimensional object, and the vertex coordinates and the line segment information between the vertices in the two-dimensional line drawing. S102: Dividing the two-dimensional line drawing according to the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices, to obtain at least one line drawing sub-picture.

S103 : 根据每一个所述线画图子图重建一个对应的三维子物体。 S103: Reconstruct a corresponding three-dimensional sub-object according to each of the line drawing sub-pictures.

S104: 组合全部的所述三维子物体得到三维物体。根据本发明实施例的三维物体的重建方法，可以对由平面组成的物体的线画图进行合理的分割和重建，且二维线画图的输入方式较为简便。本专利的一个实施例旨在解决根据获得的二维线画图，在缺少深度信息的情况下，如何自动重建三维物体的技术问题。具体地说，本专利的实施例通过一种方法把线画图按物体的自由度进行分割，然后把分割之后的子图分别重建为子物体，再把这些子物体按面、线、点结合的方式合并得到最终完整的三维物体。由于分割之后的线画图比原线画图简单得多，因此从分割后的子图分别重建三维子物体容易得到全局最优结果，更加符合人们视觉感知规律。这种方式可以解决现有技术中对复杂线画图重建三维物体时经常陷入局部最优，从而无法得到正确的三维重建物体的问题。同时，这种分割、重建、组合三维物体的方式可以正确重建复杂的三维物体。 S104: Combine all the three-dimensional sub-objects to obtain a three-dimensional object. According to the reconstruction method of the three-dimensional object according to the embodiment of the present invention, the line drawing of the object composed of the plane can be reasonably segmented and reconstructed, and the input mode of the two-dimensional line drawing is relatively simple. One embodiment of the present patent is directed to solving the technical problem of how to automatically reconstruct a three-dimensional object in the absence of depth information based on the obtained two-dimensional line drawing. Specifically, the embodiment of the present patent divides the line drawing according to the degree of freedom of the object by a method, and then reconstructs the divided sub-pictures into sub-objects respectively, and then combines the sub-objects into planes, lines, and points. The way merges to get the final complete 3D object. Since the line drawing after segmentation is much simpler than the original line drawing, it is easy to obtain the global optimal result by reconstructing the three-dimensional sub-objects from the divided sub-graphs, which is more in line with people's visual perception rules. This method can solve the problem that the prior art often falls into local optimum when reconstructing a three-dimensional object for a complex line drawing, so that the correct three-dimensional reconstruction object cannot be obtained. At the same time, this way of segmenting, reconstructing, and combining three-dimensional objects can correctly reconstruct complex three-dimensional objects.

在本发明的一个实施例中，如图 2-图 7所示，利用物体的自由度，对线画图进行符合人们视觉感知的分割，其分割结果为表示三维子物体的线画图子图，对分割后的表示基本子物体的子线画图进行三维重建，然后组合重建后的三维子物体得到重建的三维物体。 In an embodiment of the present invention, as shown in FIG. 2 to FIG. 7, the line drawing is subjected to segmentation conforming to people's visual perception by using the degree of freedom of the object, and the segmentation result is a line drawing subgraph representing the three-dimensional sub-object, The divided sub-line representation representing the basic sub-object is three-dimensionally reconstructed, and then the reconstructed three-dimensional sub-object is combined to obtain a reconstructed three-dimensional object.

S21 : 把线画图分解为表示基本子物体的线画图子图。线画图是由直线线段组合而成的平面图。物体的自由度是指确定一个三维物体所需要的最少的已知 3D 坐标的顶点个数。 S21: Decompose the line drawing into a line drawing subgraph representing the basic sub-object. A line drawing is a plan view of a combination of straight line segments. The degree of freedom of an object is the number of vertices that determine the minimum known 3D coordinates required for a three-dimensional object.

在本发明的一个实施例中，自由度是确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数。三维物体可以是物体的三维多边形表示。三维物体既可以是实际存在的物体，也可以是虚拟的图形化表示。在本发明的一个实施例中，物体可以是含有多个流形的固体，或者是单一流形的固体，或者是非固体。其中，多个流形可以是固体的一个子集。例如：复杂的房子是一个固体，房子可以由栅栏、烟筒、窗户等流形组成。 In one embodiment of the invention, the degree of freedom is to determine a three-dimensional object consisting of planes The number of minimum vertices that need to be known for spatial location. A three-dimensional object can be a three-dimensional polygonal representation of an object. A three-dimensional object can be either an actual object or a virtual graphical representation. In one embodiment of the invention, the object may be a solid containing a plurality of manifolds, or a solid of a single manifold, or a non-solid. Among them, the plurality of manifolds may be a subset of the solids. For example: A complex house is a solid, and the house can be made up of manifolds such as fences, chimneys, and windows.

流形的一般定义如下：流形，或者更准确的说是 2D流形，是局部具有欧几里得空间性质的空间，是欧几里得空间中的曲线、曲面等概念的推广。欧几里得空间就是最简单的流形的实例，地球表面这样的球面则是一个稍微复杂的例子，一般的流形可以通过把许多平直的片折弯并粘连而成。流形在数学中用于描述几何形体，它们为研究形体的可微性提供了一个自然的平台。物理上，经典力学的相空间和构造广义相对论的时空模型的四维伪黎曼流形都是流形的实例。位形空间中也可以定义流形，环面就是双摆的位形空间。所需要的最少是指重建一个三维物体所需要的顶点集合中使用的顶点数量最少。 The general definition of manifolds is as follows: manifolds, or more accurately 2D manifolds, are spaces with local Euclidean space properties, and are generalizations of concepts such as curves and surfaces in Euclidean space. The Euclidean space is an example of the simplest manifold. A spherical surface such as the Earth's surface is a slightly more complicated example. A general manifold can be formed by bending and adhering a number of straight sheets. Manifolds are used in mathematics to describe geometric shapes, which provide a natural platform for studying the variability of shapes. Physically, the phase space of classical mechanics and the four-dimensional pseudo-Riemannian manifold of the spatio-temporal model constructing general relativity are examples of manifolds. The manifold can also be defined in the configuration space, and the torus is the spatial space of the double pendulum. The minimum required is the minimum number of vertices used in the set of vertices needed to reconstruct a 3D object.

已知 3D坐标的顶点是指顶点的三维坐标（x、 y、 z坐标）都可以获知的顶点。在本专利中，三维物体的顶点是三维物体的边相交的交点；线画图的顶点是所述线画图表示的三维物体的顶点在所述线画图平面上的投影点（注：也就是线画图的顶点肯定是所述线画图的线段的交点，但是所述线画图的线段的交点不一定是所述线画图的顶点）；图（graph) 的顶点则是所述图建构上的一个固定点（图是由这些顶点和顶点之间的连线组成的结构）。通常在本专利中，在讲三维物体的时候使用的 "顶点"一词表示的是三维物体的顶点，在讲线画图的时候用的 "顶点"一词表示的是线画图的顶点，在讲图（结构）的时候用的 "顶点"一词表示的是图（结构）的顶点，如有可能混淆的时候本专利会使用完整的名词来表示。三维物体的顶点可以包括确定顶点和不确定顶点。确定顶点是已知（X、 y、 z) 坐标的顶点，确定顶点可以是已知 3D坐标的顶点，不确定顶点是顶点 3D坐标不能完全确定的顶点。例如，对于一个立方体，该立方体的自由度为 4，即只要知道该立方体的 4个不共面顶点的 3D坐标即可以根据立体解析几何的方法确定该立方体的 3D 模型。所需要的最少的已知 3D坐标的顶点个数是指重建一个由平面围成的三维物体所需要知道的最少顶点数量，并不意味着随便选择最少数量的顶点就一定可以确定该三维物体。在上述例子中，如果只知道同一平面内的 4个顶点的坐标并不能够确定立方体；但是若知道该立方体的 4个不在同一平面上的，则可以确定立方体。在本发明的一个实施例中，基本物体（或基本子物体，强调子物体主要是想说明这些物体时分割之后的结果）为自由度为 4 的物体。如图 4 (c ) 所示的物体为基本物体。 It is known that the vertices of 3D coordinates refer to vertices in which the three-dimensional coordinates (x, y, z coordinates) of the vertices are known. In this patent, the vertices of the three-dimensional object are the intersections of the intersections of the edges of the three-dimensional object; the vertices of the line drawing are the projection points of the vertices of the three-dimensional object represented by the line drawing on the line drawing plane (Note: that is, the line drawing The vertex of the line is definitely the intersection of the line segments of the line drawing, but the intersection of the line segments of the line drawing is not necessarily the vertices of the line drawing; the vertices of the graph are a fixed point on the construction of the figure (The graph is a structure consisting of the lines between these vertices and vertices). Usually in this patent, the word "vertex" used in the description of a three-dimensional object represents the apex of a three-dimensional object, and the word "vertex" used in the drawing of a line indicates the apex of a line drawing. The term "vertex" used in the diagram (structure) refers to the vertices of the graph (structure). If there is a possibility of confusion, the patent uses a complete noun. The vertices of the three-dimensional object may include determining vertices and uncertain vertices. Determine the vertices as the vertices of the known (X, y, z) coordinates, determine that the vertices can be vertices of known 3D coordinates, and the indeterminate vertices are vertices whose vertex 3D coordinates cannot be completely determined. For example, for a cube, the cube has a degree of freedom of 4, that is, as long as the 3D coordinates of the four non-coplanar vertices of the cube are known, the 3D model of the cube can be determined according to the method of stereo analytic geometry. The minimum number of vertices of the known 3D coordinates required is the minimum number of vertices needed to reconstruct a three-dimensional object enclosed by a plane, and does not mean that the three-dimensional object can be determined by randomly selecting the minimum number of vertices. In the above example, if only the coordinates of the four vertices in the same plane are known, the cube cannot be determined; but if it is known that the four cubes are not on the same plane, the cube can be determined. In one embodiment of the present invention, the basic objects (or basic sub-objects, the emphasis sub-objects are mainly intended to account for the results after the division of these objects) are objects with a degree of freedom of four. The object shown in Figure 4 (c) is the basic object.

自由度为 4的物体，是指确定该物体所需要的最少的已知 3D坐标的顶点个数为 4。在本发明的一个实施例中，通过将两个基本物体的两个面粘合在一起就得到自由度为 5的物体，在本专利中通过把一个基本物体和一个单独的（线段围城的）面（该面不和所述基本物体的任何一个面在同一个平面上）的边粘合在一起也得到一个自由度为 5的物体；将两个基本物体的两个边粘合在一起可以得到自由度为 6的物体；将两个基本物体的两个顶点粘合在一起可以得到自由度为 7的物体。可以理解的是，上述举例只是为了帮助解释自由度的概念而做出的具体示例，并不代表对本发明实施例的具体限制。 An object with a degree of freedom of 4 means that the number of vertices of the least known 3D coordinates required to determine the object is 4. In one embodiment of the invention, an object having a degree of freedom of 5 is obtained by bonding two faces of two basic objects together, in this patent by placing a basic object and a separate (line segmented) Bonding the sides of the face (which is not on the same plane as any of the basic objects) also results in an object with a degree of freedom of 5; bonding the two sides of the two basic objects together Obtain an object with a degree of freedom of 6; bonding the two vertices of the two basic objects together to obtain an object with a degree of freedom of 7. It is to be understood that the above examples are only specific examples to help explain the concept of degrees of freedom, and do not represent specific limitations of the embodiments of the invention.

如图 4所示，首先，接收线画图信息，这些信息包括：线画图所有顶点的 x、 y坐标，顶点之间的线段，以及由顶点和线段计算得到的面信息。线画图信息可以是根据用户实时输入的线画图生成的，也可以是从其他设备处获取的。例如线画图可以是由鼠标或者电脑笔在屏幕上绘出的草图，也可以是用笔在纸上绘的图，也可以是相机拍摄的线画图草图，建筑照片等等 (但不局限于这些输入方式），线画图信息是线画图的数字化表示。在本发明的一个实施例中，一个线画图可以用一个拥有边和顶点的二维图表示。用户或者设计者可以使用电脑笔、鼠标、手指（在一个触摸屏上）等工具在计算机或者移动终端的屏幕上绘出该拥有边和顶点的二维图，用户或者设计者可以绘出全部或者部分的边或顶点。在得到输入线画图之后，我们可以通过计算得到线画图的面，然后再用所有两个相邻面进行扩展，得到一个表示基本物体的线画图子图。在本发明的一个实施例中，可通过下面的计算方式得到线画图的面：线画图的面的计算方法如下：（1 ) 首先使用深度优先算法找到所有潜在面（潜在面的定义如下：它是一个环，这个环的边不相交并且环的两个非相邻顶点之间没有一条边相连）。 ( 2 ) 然后由这些潜在面构造出一个加权图，图的顶点表示一个潜在面，顶点的权值是该潜在面的边数；如果两个潜在面（顶点）可以共存（两个潜在面可以共存的条件是：如果它们没有公共边或者它们的公共边在同一直线上），那么这两个顶点之间有连线连接。（3 ) 利用求最大权值团问题的算法找到最优解，这个解便为该线画图所表示的物体的面。 As shown in FIG. 4, first, line drawing information is received, which includes: x, y coordinates of all vertices of the line drawing, line segments between vertices, and face information calculated by vertices and line segments. The line drawing information may be generated according to a line drawing input by the user in real time, or may be obtained from other devices. For example, a line drawing can be a sketch drawn on the screen by a mouse or a computer pen. It can be a picture drawn on paper with a pen, a line drawing sketch taken by a camera, an architectural picture, etc. (but not limited to these input methods), and line drawing information is a digital representation of a line drawing. In one embodiment of the invention, a line drawing can be represented by a two-dimensional map of edges and vertices. The user or designer can use a computer pen, mouse, finger (on a touch screen) and other tools to draw a two-dimensional map of the edge and vertex on the screen of the computer or mobile terminal, the user or the designer can draw all or part of The edge or apex. After getting the input line drawing, we can calculate the face of the line drawing by calculation, and then expand with all two adjacent faces to get a line drawing subgraph representing the basic object. In one embodiment of the present invention, the face of the line drawing can be obtained by the following calculation method: The face of the line drawing is calculated as follows: (1) First, all potential faces are found using the depth first algorithm (the potential face is defined as follows: Is a ring, the edges of the ring do not intersect and there is no edge between the two non-adjacent vertices of the ring). (2) Then construct a weighted graph from these potential faces. The vertex of the graph represents a potential face, and the weight of the vertex is the number of sides of the potential face; if two potential faces (vertices) can coexist (two potential faces can The conditions for coexistence are: If they have no common edges or their common edges are on the same line, then there is a wire connection between the two vertices. (3) Find the optimal solution by using the algorithm for finding the maximum weight group problem, which is the face of the object represented by the line drawing.

具体的如何计算得到线画图的面可参考 J. Liu and X. Tang. Evolutionary search for faces from line drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 7(6):861-872, 2005.中提供的方式。例如我们通过图 9 中的线画图，可以得到四个潜在面： f_:i (其顶点依次为3， b, C)、 f₂ (其顶点依次为 b, d)、 f_s (其顶点依次为^ d， C), 其顶点依次为 b， d, c)。然后根据四个面的共存关系建立权值图（如图 9(b)所示）。最后利用求解最大权值团问题的算法可以得到 fi、 f₂、 f₃、 ^四个顶点（图 9(b) 中的顶点）以及它们之间的连线组成的最大权值团是需要的解。所以在这个例子中我们最后得到四个面为：、 f₂、 f₃、 f₄。在本发明的一个实施例中，可以采用下面的方式进行相邻面的扩展：算法 1 : For details on how to calculate the line drawing, refer to J. Liu and X. Tang. Evolutionary search for faces from line drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 7(6): 861-872, 2005. The way it is provided. For example, we can get four potential faces through the line drawing in Figure 9: f _:i (the vertices are 3, b, C in order), f ₂ (the vertices are b, d in order), and f _s (the vertices are in turn For ^ d, C), the vertices are b, d, c). Then, based on the coexistence relationship of the four faces, a weight map is created (as shown in Fig. 9(b)). Finally, using the algorithm for solving the maximum weight group problem, we can get the four vertices of fi, f ₂ , f ₃ , ^ (the vertices in Fig. 9(b)) and the maximum weight group composed of the lines between them. solution. So in this example we end up with four faces: , f ₂ , f ₃ , f ₄ . In one embodiment of the invention, the expansion of adjacent faces can be performed in the following manner: Algorithm 1:

( 1 ) 首先，初始设置一个空的集合 F_teit，把还没有选择过的两个相邻的面标定为确定面并加入到集合 F_{½ t} 中，把线画图其它所有的面标定为不确定面。 (1) First, an empty set F _{teit is} initially set, two adjacent faces that have not been selected are calibrated as deterministic faces and added to the set F _{1⁄2 t} , and all other faces of the line drawing are marked as uncertain. surface.

在本发明的一个实施例中，可以预先通过假设两个相邻的面的顶点的 z 坐标已知的方式把两个相邻的面标定为确定面。其中，确定面是至少有三个不在同一直线上的确定顶点的面，不确定面是在三维空间的位置不能确定的面。检査一个不确定面 f，如果它的顶点中至少有三个不共线，且它们都属于 !^ 中的面的顶点，则根据几何性质， f的所有顶点坐标都可以被计算出来。把 f标定为确定面并加入到 ^中。根据几何性质，两点可以确定一条直线，三个不共线的点可以确定一个平面。对于一个不确定面 f，如果检査判断出它的顶点中有三个顶点不共线，且都属于 ^中的面的顶点，则这三个顶点是确定顶点；根据上述判断结果将不确定面 f标定为确定面，并加入到集合 F_{i! t}中。重复歩骤（2)，直到不存在任何一个可以加入 F_in 的不确定面。歩骤（3 ) 可以通过遍历所有的不确定面来完成，直到找不到任何一个可以加入?_½½集合的面为止。经过上述歩骤，可以得到由中所有面组合而成的表示一个基本物体的线画图子图。重复以上（1 )、（2 )、 (3 ) 歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次，从而可以得到全部的线画图子图。在得到所有可能的集合之后，通过以下算法确定最后如何分割初始线画图为线画图子图： In one embodiment of the invention, two adjacent faces may be previously calibrated as a definite surface by assuming that the z-coordinates of the vertices of two adjacent faces are known. Wherein, the determining surface is a surface having at least three determined vertices not on the same straight line, and the indeterminate surface is a surface that cannot be determined in the position of the three-dimensional space. Examine an indeterminate surface f. If at least three of its vertices are not collinear, and they all belong to the vertices of the faces in !^, all vertex coordinates of f can be calculated according to the geometric properties. Mark f as the definite surface and add it to ^. According to the geometric properties, two points can determine a straight line, and three non-collinear points can determine a plane. For an indeterminate surface f, if the check determines that three vertices in its vertices are not collinear, and both belong to the vertices of the faces in ^, then the three vertices are the determined vertices; f is calibrated to determine the face and is added to the set F _{i! t} . Repeat step (2) until there is no uncertainty surface that can be added to F _in . Ho step (3) can be done by traversing all the face of uncertainty, until not find any can _join? ½½ until the surface of the collection. After the above steps, a line drawing sub-picture representing a basic object composed of all the faces in the middle can be obtained. The above steps (1), (2), and (3) are repeated until all adjacent faces in the faces of all the two-dimensional line drawings are selected once, so that all the line drawing sub-pictures can be obtained. After getting all possible sets, determine how to finally split the initial line drawing into a line drawing subgraph by the following algorithm:

算法 2: ( 1 ) 设置集合 ^₅和5 ^^为空集。 Algorithm 2: (1) Set the collection ^ ₅ and 5 ^^ to be an empty set.

(2) 对于每一对（两个）相邻的面，使用前一个实施例提供的算法 1得到一个组成子物体的面的集合 F _i，如果得到的 F_iKit包含两个以上的面，则把这个 F_i!3i_t加入到集合 _EFS中。 (2) For each pair (two) of adjacent faces, using the algorithm 1 provided in the previous embodiment to obtain a set F _i of faces constituting the child object, if the obtained F _iKit contains more than two faces, Add this F _i!3 i _t to the set _EFS .

(3 ) 用 ^中的元素创建共存关系图 G 。如果两个物体没有共享任何一个平面或者它们的公共面只存在于同一个平面上，我们就说这两个物体具有共存关系。共存关系图 G。_¾是用来表示线画图中的物体的共存关系。共存关系图是由顶点和边组成的图结构，每一个分割得到的子线画图是 G.。^中的一个顶点，一个子图所具有的面的个数做为该顶点的权值。如果任意两个顶点所表示的物体具有共存关系，就用一条边连接这两个顶点。这样，就可以得到一个完整的共存关系图 G。_¾ 。 (3) Create a coexistence relationship graph G with the elements in ^. If two objects do not share any of the planes or their common faces exist only on the same plane, we say that the two objects have a coexistence relationship. Coexistence relationship graph G. _3⁄4 is used to represent the coexistence of objects in the line drawing. The coexistence relationship graph is a graph structure composed of vertices and edges, and the sub-line drawing obtained for each segmentation is G. A vertex in ^, the number of faces a subgraph has as the weight of the vertex. If the objects represented by any two vertices have a coexistence relationship, connect the two vertices with one edge. In this way, a complete coexistence graph G can be obtained. _3⁄4 .

(4)从 G。_¾中找到最大权值团（maximum weight clique), 并把这个最大权值团中顶点所对应的子图加入到 S_Mwe中。团是一个拥有顶点和边的图，其中的任意两个顶点之间都有一条边连接。最大权值团是所有团中权值之和最大的团。 (4) From G. Find a maximum priority value _¾ group (maximum weight clique), and the sub-picture the maximum priority value corresponding to the groups added to the vertex of the S _Mwe. A group is a graph with vertices and edges, and any two vertices have an edge connection between them. The maximum weight group is the group with the largest sum of weights in all groups.

(5 )对于没有被包含在 S_MWi；的子图中的面，根据原线画图，把它们连接到 S_Mwe中与之相邻接的子图中。在本发明的一个实施例中，对于某些特殊的线画图，可能存在没有包含在3^^集合里的面的组合，这些线画图含有的连接的面的个数多于 3个。在这种情况下，仍然可以用算法 2对它进行分割，但算法 1和算法 2要做微小的改变。在算法 1的歩骤（1 ) 中， "把两个相邻的面标定为确定面"改为 "把三个相邻的面标定为确定面"；在算法 2的（2) 中， "对于每一对相邻的面，使用算法 1得到一个组成子物体的面的集合 _ϊ½，如果得到的 F_¾ 包含两个以上的面，则把这个 F_iKdt加入到集合 S_M:EFS中" 改为 "对于每三个相邻的面，使用算法 1得到一个组成子物体的面的集合 F _ii，如果得到的包含三个以上的面，则把这个 F_¾ 加入到集合 S_M: 中 "。 (5) For the faces that are not included in the sub-picture of S _MWi ;, according to the original line drawing, connect them to the sub-picture adjacent to the S _Mwe . In one embodiment of the present invention, for some special line drawing, there may be a combination of faces that are not included in the 3^^ set, and these line drawings contain more than three connected faces. In this case, it can still be split by Algorithm 2, but Algorithm 1 and Algorithm 2 have to make minor changes. In the step (1) of Algorithm 1, "to calibrate two adjacent faces to a certain face" is changed to "mark three adjacent faces as a defined face"; in (2) of Algorithm 2, " for each pair of adjacent surfaces, using an algorithm to obtain a set consisting of a sub _ϊ½ object plane, if F _¾ obtained contains more than two faces, put the _F iKdt added to the set S _{M: EFS} of "change For "for every three adjacent faces, use Algorithm 1 obtains a set F _ii of faces that make up the sub-object. If more than three faces are obtained, this F _{3⁄4 is} added to the set S _M:.

通过上述算法 1或算法 2，可以有效的对得到的线画图的面进行扩展，从而得到分割后的线画图的子图。 By the above algorithm 1 or algorithm 2, the plane of the obtained line drawing can be effectively expanded, thereby obtaining a sub-picture of the divided line drawing.

S22: 从每一个线画图子图重建对应的三维基本子物体。 S22: Reconstruct the corresponding three-dimensional basic sub-object from each line drawing sub-picture.

通过歩骤 S21 , 把一个线画图分割为了多个表示物体的线画图子图，每一个子图都可以被看做是一个非常简单的线画图。现在可以利用最小化目标函数出线画图子图中所有顶点的

深度值。其中 Φ^' ζ₂ .. ¾_ν)的建立使用了 MSDA( Minimum Standard Deviation of Angles,角的最小标准偏差：)、面的平面性、线的平行性、等距性、角的正交性五条规则，它们的解释如下： MSDA规则要求三维物体在所有顶点上的角的标准偏差最小；面的平面性规则要求在一个面上的所有顶点在三维空间上共平面；线的平行性规则要求在线画图上平行的一对直线在三维空间中也平行；等距性规则要求三维空间中物体的边的比率与它们在线画图上的比率相同；角的正交性要求，如果一个由三条线形成的角在线画图上满足正交的角条件，它在三维空间中的三个角正交。是这个子物体的顶点个数， _Zi By step S21, a line drawing is divided into a plurality of line drawing sub-pictures representing objects, and each sub-picture can be regarded as a very simple line drawing. Now you can use the minimized objective function to draw all the vertices in the subgraph.

Depth value. Where Φ^' ζ ₂ .. 3⁄4 _ν ) is established using MSDA ( Minimum Standard Deviation of Angles), planarity of planes, parallelism of lines, equidistance, orthogonality of angles The five rules, which are explained as follows: The MSDA rule requires that the standard deviation of the angle of the three-dimensional object at all vertices is minimal; the planarity rule of the surface requires that all vertices on one face are coplanar in three-dimensional space; the parallelism rule of the line requires A pair of parallel lines on the line drawing are also parallel in three dimensions; the equidistance rule requires that the ratio of the edges of the objects in the three-dimensional space be the same as their ratio on the line drawing; the orthogonality of the angle requires that if one is formed by three lines The angular angle on the line drawing satisfies the orthogonal angular condition, which is orthogonal to the three angles in the three-dimensional space. Is the number of vertices of this child, _Zi

(i = 1,2_S N_V) 表示第 i个顶点的深度座标值， C 是规则函数的个数。是一组给定的权值，这组权值是通过实验得到，例如可以使用 100, 1, 20, 15, 20 分别对应以上五个规则。可以理解的是，上述示例只是为了帮助理解本发明的技术方案，而不应该被视为对本发明技术方案的具体限制。 ½(¾,¾,… _S ¾J的定义可以使用文献 [K. Shoji, K. Kato. And F. Toyama. 3-D interpretation of single line drawings based on entropy minimization principle. IEEE Conference on Computer Vision and Pattern Recognition (CVP ), 2001]、 [y. Leclerc and M. Fischler. An Optimization-Based Approach to the Interpretation of Single Line. Drawings as 3D Wire Frames. International Journal of Computer Vision (IJCV), 9(2): 113-136, 1992]、 [H. Lipson and M. Shpitaln. optimization based reconstruction of a 3d object from a single freehand line drawing. Computer-Aided Design, 28(7):651-663. 1996]、 [A. Turner, D. Chapman, and A. Penn. Sketching space. Computer and Graphics, 24:869-879, 2000]、 [J. Liu, L. Cao, Z. Li and X. Tang. Plane-based optimization for 3D object reconstruction from single line drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 30(2):315-327, 2008]中提到的规则函数，比如夹角的标准方差等函数，，是这个子物体的顶点个数， (i = 1,2, 表示第 i个顶点的深度座标值，可以是文献 [K. Shoji, K. Kato. And F. Toyama. 3-D interpretation of single line drawings based on entropy minimization principle. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2001]、 [y. Leclerc and M. Fischler. An Optimization-Based Approach to the Interpretation of Single Line. Drawings as 3D Wire Frames. International Journal of Computer Vision (IJCV), 9(2): 113-136, 1992]、 [H. Lipson and M. Shpitaln. optimization based reconstruction of a 3d object from a single freehand line drawing. Computer-Aided Design, 28(7):651-663. 1996]、 [A. Turner, D. Chapman, and A. Penn. Sketching space. Computer and Graphics, 24:869-879, 2000]、 [J. Liu, L. Cao, Z. Li and X. Tang. Plane-based optimization for 3D object reconstruction from single line drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 30(2):315-327, 2008]等中提到的权值， C是规则函数的个数。 (i = 1,2 _S N _V ) represents the depth coordinate value of the ith vertex, and C is the number of regular functions. It is a given set of weights. This set of weights is obtained experimentally. For example, 100, 1, 20, 15, 20 can be used to correspond to the above five rules. It can be understood that the above examples are only for helping to understand the technical solution of the present invention, and should not This is considered to be a specific limitation of the technical solution of the present invention. 1⁄2(3⁄4,3⁄4,... _S 3⁄4J can be defined using the literature [K. Shoji, K. Kato. And F. Toyama. 3-D interpretation of single line drawings based on entropy minimization principle. IEEE Conference on Computer Vision and Pattern Recognition (CVP), 2001], [y. Leclerc and M. Fischler. An Optimization-Based Approach to the Interpretation of Single Line. Drawings as 3D Wire Frames. International Journal of Computer Vision (IJCV), 9(2): 113- 136, 1992], [H. Lipson and M. Shpitaln. optimization based reconstruction of a 3d object from a single freehand line drawing. Computer-Aided Design, 28(7): 651-663. 1996], [A. Turner, D. Chapman, and A. Penn. Sketching space. Computer and Graphics, 24: 869-879, 2000], [J. Liu, L. Cao, Z. Li and X. Tang. Plane-based optimization for 3D object reconstruction From the single line drawings. IEEE rules on Pattern Analysis and Machine Intelligence (TPAMI), 30(2): 315-327, 2008], the rule functions, such as the standard deviation of the angle, etc., are the sub-objects. The number of points, (i = 1, 2, represents the depth coordinate value of the ith vertex, which can be the literature [K. Shoji, K. Kato. And F. Toyama. 3-D interpretation of single line drawings based on entropy IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2001], [y. Leclerc and M. Fischler. An Optimization-Based Approach to the Interpretation of Single Line. Drawings as 3D Wire Frames. International Journal of Computer Vision (IJCV), 9(2): 113-136, 1992], [H. Lipson and M. Shpitaln. optimization based reconstruction of a 3d object from a single freehand line drawing. Computer-Aided Design, 28(7):651 -663. 1996], [A. Turner, D. Chapman, and A. Penn. Sketching space. Computer and Graphics, 24: 869-879, 2000], [J. Liu, L. Cao, Z. Li and X Tang. Plane-based optimization for 3D object reconstruction from single line drawings. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 30(2): 315-327, 2008], etc. The number of functions.

S23 :通过点、线、面结合的方式把重建的子物体结合为完整的三维物体。由于在歩骤 S22 中各个子物体是分别重建的，所以在合并它们的时候需要通过原线画图得到各个子物体的结合部位信息，同时通过比较结合部位的尺寸重新调整需要结合的两个部分的大小，把所有子物体结合为整体之后，再根据最小化目标函数 , ¾ 调整整个物体的大小，这里表示整个物体的顶点个数。可以理解的是，基本子物体的结合可以是点点处结合，也可以是点线处结合。例如需要结合的两个子物体可能结合部位是第一个物体的一个点和第二个物体也的一个点上；也可能是第一个物体的一个点和第二个物体的一条线上某个位置；也可能是第一个物体的一个点和第二个物体的一个面上某个位置; 也可能是第一个物体的一条线和第二个物体的一条线上某个位置；也可能是第一个物体的一条线和第二个物体的一个面上某个位置；也可能是第一个物体的一个面和第二个物体的一个面。下面结合附图和具体场景中的实施例对本专利进一歩说明。在本发明的一个实施例中，当我们想到一 -水个物体并需要快速展现给其他人看时，或者需要通过搜索找到类似的物体的时候，我们可以在触屏设备上快速画出一个线画图（如图 4 (a) 所示），触屏设备可以是平板电脑、手机等。在输入线画图之后，触屏设备可以得到对应的物体的顶点的 X， y坐标和连接这些顶点的线条（如图 4 (b)所示）。在图 4 (b) 中，得到的线画图对应的顶点分别是 ( ^,?)，（¾, y₂ ? )， ……， (x₇j₇J.)，我们把这些顶点分别命名为：S23: Combine the reconstructed sub-objects into a complete three-dimensional object by means of point, line and surface combination. Since each sub-object is reconstructed separately in step S22, it is necessary to obtain the joint position information of each sub-object through the original line drawing, and re-adjust the two parts to be combined by comparing the size of the joint portion. Size, after combining all the sub-objects as a whole, according to the minimized objective function, 3⁄4 adjust the size of the whole object, here represents the number of vertices of the entire object. It can be understood that the combination of the basic sub-objects may be a combination of points or a combination of dotted lines. For example, two sub-objects that need to be combined may be a point of the first object and a point of the second object; or a point of the first object and a line of the second object Position; it may be a point of the first object and a position on one side of the second object; it may also be a line of the first object and a position on a line of the second object; It is a line of the first object and a position on one side of the second object; it may also be a face of the first object and a face of the second object. The present patent will be further described below in conjunction with the drawings and the embodiments in the specific scenarios. In one embodiment of the present invention, when we think of a water object and need to quickly show it to others, or need to find a similar object by searching, we can quickly draw a line on the touch screen device. Drawing (as shown in Figure 4 (a)), the touch screen device can be a tablet, a mobile phone, and the like. After the input line is drawn, the touch screen device can obtain the X, y coordinates of the vertices of the corresponding object and the lines connecting the vertices (as shown in Figure 4 (b)). In Figure 4 (b), the corresponding vertex corresponding to the line drawing is ( ^,?), (3⁄4, y ₂ ? ), ..., (x ₇ j ₇ J.), we named these vertices as :

V, , V，。其中 "？"表示暂时还无法确定的参数，其它诸如¾，符号表示已经知道顶点坐标的数值。然后可以根据算法 1或算法 2描述的方法分割线画图得到图 4 (c ) 中的两个分割好的子图。具体地说，通过以下歩骤实施：令禾口 S:MW'C为空集；对于每一对相邻的面，使用算法 1得到一个组成子物体的面的集合 F_TEIT，如果得到的 F_IN½包含两个以上的面，则把这个 F_INIT加入到集合 S_MEFS中；用 S_MEFS中的元素创建共存关系图 G。_¾；从^< ^中找到最大权值团（maximum weight clique ) , 并把这个最大权值团中顶点所对应的子图加入到¾«^中；对于没有被包含在 S_M 的子图中的面，根据原始输入的线画图，把它们连接到 S_Mwe中与之相邻接的子图中。在从线画图中分割出线画图子图之后，再对每个分割好的子图分别重建三维物体，计算出图 4 (b ) 中之前无法确定的参数值。具体的说，通过以下歩骤实施：其中

N_v是这个子物体的顶点个数， ¾ ( 1 = 1._;2,， N_v)表示第 i个顶点的深度座标值，

S22 中提到的规则函数，可以是歩骤 S22中提到的权值， C是规则函数的个数。最后重新结合这些重建的三维子物体，得到一个完整的三维物体。具体地说，我们可以通过以下歩骤结合所有重建好的基本物体：在合并它们的时候需要通过原线画图得到各个子物体的结合部位信息，同时通过比较结合部位的尺寸重新调整需要结合的两个部分的大小，把所有子物体结合为一个整体之后，再根据最小化目标函数 Φ(¾, ,，¾^重新调整一次整个物体的大小，这里 Ν_ν是整个物体的顶点个数。在本发明的另一个实施例中，线画图的输入方式亦可不借助手机或平板电脑等触屏设备的手写输入，而是通过摄像头拍摄或者扫描仪扫描在图纸上的线画图的方式获得线画图。 V, , V, . Where "?" indicates a parameter that cannot be determined for the time being, and other such as 3⁄4, the symbol indicates that the value of the vertex coordinates is already known. The two divided subgraphs in Figure 4(c) can then be obtained by dividing the line drawing according to the method described in Algorithm 1 or Algorithm 2. Specifically, it is implemented by the following steps: Let S and MW'C be empty sets; for each pair of adjacent faces, use Algorithm 1 to obtain a set of faces F _TEIT that make up the sub-objects, if the obtained F _IN1⁄2 contains more than two faces, then this F _{INIT is} added to the set S _MEFS ; Create a coexistence relationship graph G with elements in S _MEFS . _3⁄4 ; Find the maximum weight clique from ^< ^, and add the subgraph corresponding to the vertex in this maximum weight group to 3⁄4«^; for the subgraph that is not included in S _M Faces, based on the original input line drawing, connect them to the sub-pictures in the S _Mwe that are adjacent to them. After the line drawing sub-picture is segmented from the line drawing, the three-dimensional object is reconstructed for each of the divided sub-pictures, and the parameter values that cannot be determined before in FIG. 4(b) are calculated. Specifically, it is implemented through the following steps:

N _v is the number of vertices of this sub-object, 3⁄4 ( 1 = 1. _; 2,, N _v ) represents the depth coordinate value of the i-th vertex,

The rule function mentioned in S22 may be the weight mentioned in step S22, and C is the number of rule functions. Finally, these reconstructed three-dimensional sub-objects are recombined to obtain a complete three-dimensional object. Specifically, we can combine all the reconstructed basic objects by the following steps: When merging them, we need to obtain the binding part information of each sub-object through the original line drawing, and at the same time, re-adjust the two parts that need to be combined by comparing the size of the combined part. The size of the parts, after combining all the sub-objects into one, then re-adjust the size of the whole object according to the minimum objective function Φ(3⁄4, ,,3⁄4^, where Ν _ν is the number of vertices of the whole object. In another embodiment of the invention, the input method of the line drawing can also obtain the line drawing by means of camera shooting or by scanning the line drawing on the drawing without using the handwriting input of the touch screen device such as a mobile phone or a tablet computer.

如果是通过拍摄或者扫描的方式获取线画图后，首先得到照片中的点和线并提取出对应的线画图，如图 4 (b) 所示。在图 4 (b ) 中，得到的线画图对应的顶点分别是 O ^i,？)， ……，（x₇.,y₇ ?)，我们把这些顶点分别命名为：， v₂， ……， v₇。其中 "？"表示暂时无法确定的参数，其它诸如 _Xi，等符号表示已经知道数值的参数。通过前述实施例中同样的歩骤我们可以重建出三维物体。在本发明的另一个实施例中，当我们得到一个线画图之后（如图 5 (a) 所示），首先对线画图进行分割得到图 5 (b) 的两个子图，具体的分割方式可以参考歩骤 S21 中的算法。图 5 (b) 有两个分割后的子线画图，它们分别是物体 A和物体 B。物体 A的顶点分别是 al、 a2、 a3、 a4，物体 B的顶点分别是 al、 b2、 b3、 b4，然后对图 5 (b) 中的两个线画图子图分别进行重建得到图 5 (c) 中的四个可能的物体。对这四个物体在 al处结合，可以得到重建后四个可能的物体。最后可以由用户选择一种认为合适的重建结果（例如图 5 (d) )。可以理解的是，上述举例只是为了帮助理解本发明实施例而做出的一种示例，而不是对本发明实施例的具体限制，最终的重建结果也可以参考其他参数确定或者随机进行选择。在本发明的另一个实施例中，主要展现线画图子图之间的线线结合的过程。当我们得到一个线画图之后（如图 6 (a) 所示），首先对线画图进行分割得到图 6 (b) 的两个子图。图 6 (b) 有两个分割后的子图，它们分别是物体 A和物体 B。物体 A的顶点分别是 al、 a2、 a3、 a4，物体 B的顶点分别是 bl、 b2、 b3、 b4。然后对图 6 (b ) 的两个子线画图分别进行重建得到图 6 (c) 物体的四个可能形状。从图 6 (a) , 我们可以知道这个三维物体是由图 6 (c) 中上面两个物体之一的 al和 a3组成的线段和下面两个物体之一的 bl和 b2组成的线段结合在一起。在这个部位上结合图 6 (c ) 中的子物体可以得到四个可能的三维物体。最后由用户选择一种他们认为合适的重建结果（例如图 6 ( d) )。在本发明的另一个实施例中，首先我们通过照相机拍摄一幅建筑物的照片（如图 7 (a) 所示），然后把照片输入电脑，在电脑上使用鼠标等工具画出希望重建的建筑物的轮廓（如图 7 (b )所示）。在图 7 (b) 中我们使用了虚线表示被遮挡部分的轮廓线。在画完轮廓的顶点和线段之后，使用前述实施例中的算法 1或算法 2中的方法对轮廓线（线画图）进行分割。在这个例子中分割之后得到的三个表示基本物体的线画图子图（如图 7 (c) 所示）。对这些分割结果使用前述实施例中重建基本物体的方法中描述的方式重建，重建结果如图 7 (d)所示。最后通过前述点、线、面结合的方式把重建的子物体结合为完整的三维物体的实施例中描述的方法结合各个重建的结果，最后得到完整的三维物体（如图 7 (e) 所示）。通过加入建筑物的纹理就可以得到最后的三维重建结果。图 7 (f) 是从多个视角观察重建结果的示意图。本专利提出的方法使得人们得到三维模型的绘图、输入方式更加自然、工作量更小，而且普通用户也可无需特别培训就能轻松掌握；本专利提出的方法可以对任意由平面组成的物体的线画图进行合理的分割和重建；本专利提出的方法分割出来的表示三维物体的线画图子图在重建三维物体的过程中易于得到最优解。图 8是按本专利方法重建的一些三维物体的效果图。图 8给出了 9个由较为复杂的线画图重建三维物体的例子，这 9个例子的重建过程都是由按本专利提出的方法编写的计算机程序自动完成的。每个例子由 4个子图构成，其中左上的子图表示输入的线画图，右上角是分割后的表示子物体的子线画图，右下角和左下角的图是把重建后的子物体重新组合为完整的三维物体的两个不同视角的投影图。下面结合图 11所示描述根据本发明实施例的一种三维物体重建装置 110，装置 110包括：获取模块 111，用于获取三维物体的二维线画图信息，二维线画图信息包括二维线画图；分割模块 112，用于根据获取模块 111获取的所述二维线画图信息，按照自由度分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数；重建模块 113，用于根据分割模块 112得到的所述线画图子图重建三维子物体；组合模块 114，用于组合全部的重建模块 113得到的三维子物体得到三维物体。在本发明的一个实施例中，获取模块 111通过下列之一的方式获取三维物体的二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式和通过从预先存储的照片中提取线画图的方式。在本发明的一个实施例中，分割模块 112包括：面信息获取单元 1121，用于根据获取模块 111获得的二维线画图信息得到二维线画图的顶点坐标、顶点之间的线段信息，根据二维线画图的顶点坐标和所述顶点之间的线段信息得到二维线画图的面信息；面分割单元 1122，用于根据面信息获取单元 1121得到的二维线画图的面信息，按照自由度分割所述二维线画图，得到至少一个线画图子图。 If the line drawing is obtained by shooting or scanning, first obtain the points and lines in the photo and extract the corresponding line drawing, as shown in Figure 4 (b). In Figure 4 (b), the resulting line drawing The corresponding vertices are O ^i,? ), ..., (x ₇ ., y ₇ ?), we name these vertices as: , v ₂ , ..., v ₇ . Where "?" indicates a parameter that cannot be determined temporarily, and other symbols such as _Xi , etc. indicate a parameter that already knows the value. We can reconstruct a three-dimensional object by the same steps in the previous embodiment. In another embodiment of the present invention, after we obtain a line drawing (as shown in FIG. 5(a)), the line drawing is first divided to obtain two sub-pictures of FIG. 5(b), and the specific dividing manner may be Refer to the algorithm in step S21. Figure 5 (b) There are two divided sub-line drawings, which are object A and object B, respectively. The vertices of object A are al, a2, a3, a4, respectively. The vertices of object B are al, b2, b3, and b4, respectively, and then the two line drawing subgraphs in Fig. 5(b) are respectively reconstructed to obtain Fig. 5 ( c) Four possible objects. By combining these four objects at the al, you can get four possible objects after reconstruction. Finally, the user can select a reconstruction result that is deemed appropriate (eg, Figure 5(d)). It is to be understood that the above examples are only examples to help understand the embodiments of the present invention, and are not specifically limited to the embodiments of the present invention. The final reconstruction results may also be determined with reference to other parameters or randomly selected. In another embodiment of the invention, the process of wire bonding between line drawing subgraphs is primarily exhibited. When we get a line drawing (as shown in Figure 6 (a)), we first divide the line drawing to get the two sub-pictures of Figure 6 (b). Figure 6 (b) has two divided subgraphs, which are object A and object B, respectively. The vertices of the object A are al, a2, a3, and a4, respectively, and the vertices of the object B are bl, b2, b3, and b4, respectively. Then, the two sub-line drawings of Fig. 6(b) are reconstructed separately to obtain the four possible shapes of the object of Fig. 6(c). From Fig. 6(a), we can know that this three-dimensional object is composed of a line segment consisting of a1 and a3 of one of the upper two objects in Fig. 6(c) and a line segment consisting of bl and b2 of one of the following two objects. together. Four possible three-dimensional objects can be obtained by combining the sub-objects in Fig. 6(c) at this location. Finally, the user selects a reconstruction result that they think is appropriate (eg, Figure 6(d)). In another embodiment of the present invention, first, we take a picture of a building through a camera (as shown in Figure 7 (a)), then input the picture into a computer, and use a tool such as a mouse on the computer to draw a desired reconstruction. The outline of the building (as shown in Figure 7 (b)). In Figure 7(b) we use a dashed line to indicate the outline of the occluded part. After drawing the vertices and line segments of the contour, using the foregoing embodiment The method in Algorithm 1 or Algorithm 2 divides the outline (line drawing). The three line drawing subgraphs representing the basic objects obtained after the segmentation in this example (as shown in Figure 7 (c)). The reconstruction results are reconstructed in the manner described in the method of reconstructing the basic object in the foregoing embodiment, and the reconstruction result is as shown in Fig. 7(d). Finally, the method described in the embodiment of combining the reconstructed sub-objects into a complete three-dimensional object by the combination of the aforementioned points, lines and surfaces is combined with the results of the respective reconstructions, and finally a complete three-dimensional object is obtained (as shown in Fig. 7(e) ). The final 3D reconstruction result can be obtained by adding the texture of the building. Figure 7 (f) is a schematic diagram of the reconstruction results from multiple perspectives. The method proposed in this patent enables people to obtain three-dimensional model drawing, input mode is more natural, and the workload is smaller, and ordinary users can easily grasp without special training; the method proposed in this patent can be used for any object composed of planes. The line drawing is reasonably segmented and reconstructed; the line drawing subgraph representing the three-dimensional object segmented by the method proposed in this patent is easy to obtain an optimal solution in the process of reconstructing the three-dimensional object. Figure 8 is an effect diagram of some three-dimensional objects reconstructed in accordance with the method of the present patent. Figure 8 shows nine examples of reconstruction of three-dimensional objects from more complex line drawings. The reconstruction of these nine examples is done automatically by a computer program written in accordance with the method proposed in this patent. Each example consists of four subgraphs, where the upper left subgraph represents the input line drawing, the upper right corner is the divided sub-line drawing representing the sub-object, and the lower right and lower left corners are the recombination of the reconstructed sub-objects. A projection of two different perspectives of a complete three-dimensional object. A three-dimensional object reconstruction apparatus 110 according to an embodiment of the present invention is described below with reference to FIG. 11. The apparatus 110 includes: an acquisition module 111, configured to acquire two-dimensional line drawing information of a three-dimensional object, and the two-dimensional line drawing information includes a two-dimensional line. The drawing module 112 is configured to divide the two-dimensional line drawing according to the degree of freedom according to the two-dimensional line drawing information acquired by the obtaining module 111, to obtain at least one line drawing sub-picture, wherein the degree of freedom is determined by one a minimum number of vertices to be known by the spatial position of the three-dimensional object composed of planes; a reconstruction module 113, configured to reconstruct a three-dimensional sub-object according to the line drawing sub-picture obtained by the segmentation module 112; and a combination module 114 for combining All three-dimensional sub-objects obtained by the reconstruction module 113 The body gets a three-dimensional object. In an embodiment of the present invention, the acquiring module 111 acquires two-dimensional line drawing information of the three-dimensional object by one of the following methods: a manner of shooting by the imaging device, a method of input by the touch device, a manner of input by the mouse, and a pass The way in which line drawings are extracted from pre-stored photos. In an embodiment of the present invention, the segmentation module 112 includes: a surface information acquiring unit 1121, configured to obtain vertex coordinates of the two-dimensional line drawing and line segment information between the vertices according to the two-dimensional line drawing information obtained by the obtaining module 111, according to The vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices obtain the surface information of the two-dimensional line drawing; the surface dividing unit 1122 is configured to perform the surface information of the two-dimensional line drawing obtained by the surface information acquiring unit 1121, according to the freedom Dividing the two-dimensional line drawing to obtain at least one line drawing sub-picture.

在本发明的一个实施例中，面分割单元 1122具体用于：在全部的二维线画图的面中选择任意一对还没有选择过的面，作为第一线画图面和第二线画图面，其中，第一线画图面和所述第二线画图面是相邻的面；设置第一集合为空集，将第一线画图面和第二线画图面标定为确定面并加入到第一集合中；在全部的二维线画图的面中将除第一线画图面和第二线画图面的其它线画图面标定为不确定面；在不确定面中选择第三线画图面，将第三线画图面标定为确定面并加入到第一集合中，重复本歩骤直到遍历完全部的不确定面，其中，第三线画图面至少有三个不共线的顶点，且不共线的顶点是第一集合中的线画图面的顶点；根据第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到全部的二维线画图的面中的所有相邻面对都被选择过一次。 In one embodiment of the present invention, the surface segmentation unit 1122 is specifically configured to: select any pair of faces that have not been selected in the faces of all the two-dimensional line drawings, as the first line drawing surface and the second line drawing surface, The first line drawing surface and the second line drawing surface are adjacent faces; setting the first set to an empty set, and the first line drawing surface and the second line drawing surface are marked as determining faces and added to the first set ; in the face of all 2D line drawing, mark the other line drawing surfaces except the first line drawing surface and the second line drawing surface as the indeterminate surface; select the third line drawing surface in the indeterminate surface, and draw the third line drawing surface The calibration is performed to determine the face and is added to the first set, and the step is repeated until the entire face is traversed, wherein the third line drawing surface has at least three vertices that are not collinear, and the collinear vertices are the first set. The vertices of the line drawing planes; a corresponding line drawing subgraph is obtained according to the determining surface in the first set; the above steps are repeated until all the adjacent faces in the faces of all the two-dimensional line drawing are selected Once.

在本发明的另一个实施例中，面分割单元 1122还用于：如果第一集合中的确定面的数目不小于三，则将第一集合所表示的线画图子图加入到第二集合中； In another embodiment of the present invention, the face segmentation unit 1122 is further configured to: if the number of the determined faces in the first set is not less than three, add the line drawing subgraph represented by the first set to the second set ;

根据第二集合中所有的线画图子图得到共存关系图，其中，共存关系图的一个顶点表示一个线画图子图，线画图子图所具有的面的个数做为顶点的权值，如果两个线画图子图之间存在共存关系，则对应两个线画图子图的顶点之间有连线连接；共存关系图表示线画图子图所表示的物体之间的共存关系，共存关系是指两个线画图子图所表示的物体之间没有共享任何一个平面或者两个线画图子图所表示的物体之间的公共面只存在于同一个平面上； A coexistence relationship graph is obtained according to all the line drawing subgraphs in the second set, wherein one vertex of the coexistence graph represents a line drawing subgraph, and the number of faces of the line drawing subgraph is used as a vertex weight Value, if there is a coexistence relationship between the two line drawing subgraphs, there is a connection between the vertices of the corresponding two line drawing subgraphs; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing subgraphs, The coexistence relationship means that no object is shared between the objects represented by the two line drawing subgraphs or the common surface between the objects represented by the two line drawing subgraphs exists only on the same plane;

在共存关系图中找到最大权值团，确定最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，团中的任意两个顶点之间都有一条边连接，最大权值团是在共存关系图的所有团中权值之和最大的团。 Find the maximum weight group in the coexistence relationship diagram, and determine the line drawing subgraph corresponding to the vertex of the maximum weight group, where the group is a picture with vertices and edges, and there is a side between any two vertices in the group Connection, the maximum weight group is the group with the largest sum of weights in all groups of the coexistence graph.

在本发明的一个实施例中，重建模块 113包括： In one embodiment of the invention, the reconstruction module 113 includes:

重建信息获取单元 1131，用于根据最小化目标函数得到线画图子图的顶点的深度值，获取线画图子图的顶点坐标，获取三维子物体的面； The reconstruction information acquiring unit 1131 is configured to obtain a depth value of a vertices of the line drawing subgraph according to the minimum target function, obtain a vertex coordinate of the line drawing subgraph, and obtain a surface of the three-dimensional sub-object;

重建单元 1132，用于根据重建信息获取单元 1131得到的线画图子图的顶点的深度值、线画图子图的顶点坐标和三维子物体的面得到三维子物体。 The reconstruction unit 1132 is configured to obtain a three-dimensional sub-object according to the depth value of the vertices of the line drawing sub-picture obtained by the reconstruction information acquiring unit 1131, the vertex coordinates of the line drawing sub-picture, and the surface of the three-dimensional sub-object.

在本发明的一个实施例中，组合模块 114具体用于： In an embodiment of the invention, the combination module 114 is specifically configured to:

通过部位结合的方式组合全部的三维子物体，得到三维物体，部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。 The three-dimensional object is combined by means of a combination of parts to obtain a three-dimensional object, and the manner of combining the parts includes at least one of the following methods: point combination, point line combination, point surface combination, line bonding, line surface bonding, face bonding Combined with the dotted line.

在本发明的一个实施例中，组合模块 114具体用于：根据二维线画图信息得到三维子物体的结合部位信息； In an embodiment of the present invention, the combining module 114 is specifically configured to: obtain binding position information of the three-dimensional sub-object according to the two-dimensional line drawing information;

根据三维子物体的结合部位信息调整三维子物体的大小，得到调整后的三维子物体； Adjusting the size of the three-dimensional sub-object according to the information of the joint position of the three-dimensional sub-object, and obtaining the adjusted three-dimensional sub-object;

组合调整后的三维子物体得到初始三维物体； Combining the adjusted three-dimensional sub-objects to obtain an initial three-dimensional object;

根据最小化目标函数调整初始三维物体，得到三维物体。 The initial three-dimensional object is adjusted according to the minimized objective function to obtain a three-dimensional object.

下面结合图 12描述根据本发明实施例的一种三维物体重建的装置 120，装置 120包括：信息获取模块 121，用于获取三维物体的二维线画图，以及二维线画图中的顶点坐标、顶点之间的线段信息；线画图子图分割模块 122，用于根据信息获取模块 121获取的二维线画图的顶点坐标和顶点之间的线段信息分割二维线画图，得到至少一个的线画图子图；三维子物体重建模块 123，用于根据线画图子图分割模块 122得到的每一个线画图子图重建一个对应的三维子物体；三维物体组合模块 124，用于组合三维子物体重建模块 123重建的全部的三维子物体得到三维物体。在本发明的一个实施例中，线画图子图分割模块 122，包括：二维线画图的面信息获取单元 1221，根据信息获取模块 121获取的二维线画图的顶点和顶点之间的线段信息获得二维线画图的面信息；线画图子图分割单元 1222，用于根据二维线画图的面信息获取单元 1221 获取的二维线画图的面信息分割所述二维线画图，得到至少一个的线画图子图。在本发明的一个实施例中，线画图子图分割单元 1222具体用于：在全部的二维线画图的面中选择任意一对还没有选择过的面，作为第一线画图面和第二线画图面，其中，第一线画图面和第二线画图面是相邻的面；设置第一集合为空集，将第一线画图面和第二线画图面标定为确定面并加入到第一集合中；在全部的二维线画图的面中将除第一线画图面和第二线画图面的其它线画图面标定为不确定面；在不确定面中选择第三线画图面，将第三线画图面标定为确定面并加入到第一集合中，重复本歩骤直到遍历完全部的不确定面，其中，第三线画图面至少有三个不共线的顶点，且不共线的顶点是第一集合中的线画图面的顶点；根据第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到全部的二维线画图的面中的所有相邻面对都被选择过一次。在本发明的一个实施例中，如果第一集合中的确定面的数目不小于三，则线画图子图分割单元 1222还用于：将第一集合所表示的线画图子图加入到第二集合中；根据第二集合中所有的线画图子图得到共存关系图，其中，共存关系图的一个顶点表示一个线画图子图，线画图子图所具有的面的个数做为顶点的权值，如果两个线画图子图之间存在共存关系，则对应两个线画图子图的顶点之间有连线连接；共存关系图表示线画图子图所表示的物体之间的共存关系，共存关系是指两个线画图子图所表示的物体之间没有共享任何一个平面或者两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在共存关系图中找到最大权值团，确定最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，团中的任意两个顶点之间都有一条边连接，最大权值团是在共存关系图的所有团中权值之和最大的团。在本发明的一个实施例中，三维子物体重建模块 123包括：三维子物体重建信息获取单元 1231，用于根据最小化目标函数得到线画图子图的顶点的深度值，获取线画图子图的顶点坐标，获取三维子物体的面；三维子物体重建单元 1232，用于根据重建信息获取单元 1231得到的线画图子图的顶点的深度值、线画图子图的顶点坐标和三维子物体的面得到三维子 A device 120 for reconstructing a three-dimensional object according to an embodiment of the present invention will be described below with reference to FIG. The device 120 includes: an information acquiring module 121, configured to acquire a two-dimensional line drawing of the three-dimensional object, and a vertex coordinate in the two-dimensional line drawing, and line segment information between the vertices; and a line drawing sub-graph dividing module 122, configured to acquire according to the information The line segment information between the vertex coordinates of the two-dimensional line drawing and the vertex obtained by the module 121 is divided into two-dimensional line drawing to obtain at least one line drawing sub-picture; the three-dimensional sub-object reconstruction module 123 is configured to divide the module 122 according to the line drawing sub-picture. Each of the obtained line drawing subgraphs reconstructs a corresponding three-dimensional sub-object; the three-dimensional object combining module 124 is configured to combine all the three-dimensional sub-objects reconstructed by the three-dimensional sub-object reconstruction module 123 to obtain a three-dimensional object. In an embodiment of the present invention, the line drawing sub-picture segmentation module 122 includes: a two-dimensional line drawing surface information acquiring unit 1221, and line segment information between the vertex and the vertex of the two-dimensional line drawing acquired by the information acquiring module 121. Obtaining the surface information of the two-dimensional line drawing; the line drawing sub-picture dividing unit 1222 is configured to divide the two-dimensional line drawing according to the surface information of the two-dimensional line drawing acquired by the surface information acquiring unit 1221 of the two-dimensional line drawing, to obtain at least one Line drawing subgraph. In an embodiment of the present invention, the line drawing sub-graph dividing unit 1222 is specifically configured to: select any pair of faces that have not been selected in the face of all the two-dimensional line drawing, as the first line drawing surface and the second line. a drawing surface, wherein the first line drawing surface and the second line drawing surface are adjacent faces; setting the first set to an empty set, and the first line drawing surface and the second line drawing surface are marked as determining faces and added to the first set In the faces of all the two-dimensional line drawing, the other line drawing surfaces except the first line drawing surface and the second line drawing surface are marked as uncertain faces; the third line drawing surface is selected in the uncertain plane, and the third line drawing is performed The surface is calibrated to determine the surface and added to the first set. Repeat this step until the entire surface of the uncertainty is traversed, where the third line is drawn to There are three vertices that are not collinear, and the vertices that are not collinear are the vertices of the line drawing faces in the first set; a corresponding line drawing subgraph is obtained according to the determined faces in the first set; repeat the above steps until All adjacent faces in the face of all 2D line drawings are selected once. In an embodiment of the present invention, if the number of the determined faces in the first set is not less than three, the line drawing sub-graph dividing unit 1222 is further configured to: add the line drawing sub-picture represented by the first set to the second In the set; a coexistence relationship graph is obtained according to all the line drawing subgraphs in the second set, wherein one vertex of the coexistence graph represents a line drawing subgraph, and the number of faces of the line drawing subgraph is used as a vertex weight Value, if there is a coexistence relationship between the two line drawing subgraphs, there is a connection between the vertices of the corresponding two line drawing subgraphs; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing subgraphs, Coexistence relationship means that no object is shared between the objects represented by the two line drawing subgraphs or the common surface between the objects represented by the two line drawing subgraphs exists only on the same plane; in the coexistence relationship diagram Find the maximum weight group and determine the line drawing subgraph corresponding to the vertex of the maximum weight group, where the group is a picture with vertices and edges, and there is one between any two vertices in the group Edge-connected, the maximum weight in the group is the largest group of the weights of all groups coexistence and FIG. In an embodiment of the present invention, the three-dimensional sub-object reconstruction module 123 includes: a three-dimensional sub-object reconstruction information acquiring unit 1231, configured to obtain a depth value of a vertex of the line drawing sub-picture according to the minimized objective function, and obtain a line drawing sub-picture. The vertex coordinates, the surface of the three-dimensional sub-object is acquired; the three-dimensional sub-object reconstruction unit 1232, the depth value of the vertex of the line drawing sub-picture obtained by the reconstruction information acquiring unit 1231, the vertex coordinates of the line drawing sub-picture, and the surface of the three-dimensional sub-object Get three-dimensional

在本发明的一个实施例中，三维物体组合模块 124具体用于：通过部位结合的方式组合全部的三维子物体，得到三维物体，部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。在本发明的一个实施例中，三维物体组合模块 124具体用于：根据二维线画图信息得到三维子物体的结合部位信息；根据三维子物体的结合部位信息调整三维子物体的大小，得到调整后的三维子物体；组合调整后的三维子物体得到初始三维物体；根据最小化目标函数调整初始三维物体，得到三维物体。下面结合图 13描述根据本发明实施例的一种三维物体重建的终端 130，终端 130包括：三维物体重建装置 131，显示器 132，其中，显示器 132用于显示三维物体。所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的终端、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。在本申请所提供的几个实施例中，应该理解到，所揭露的终端、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。另外，在本发明各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备（可以是个人计算机，服务器，或者网络设备等）执行本发明各个实施例所述方法的全部或部分歩骤。而前述的存储介质包括： U盘、移动硬盘、只读存储器（ROM, ead-Only Memory), 随机存取存储器（RAM, Random Access Memory ) , 磁碟或者光盘等各种可以存储程序代码的介质。 In one embodiment of the present invention, the three-dimensional object combination module 124 is specifically configured to: combine all three-dimensional sub-objects by means of part combination to obtain a three-dimensional object, and combine the parts The method includes at least one of the following methods: point combination, point line combination, point plane combination, line combination, line surface combination, face combination, and point line surface combination. In one embodiment of the present invention, the three-dimensional object combination module 124 is specifically configured to: obtain the joint location information of the three-dimensional sub-object according to the two-dimensional line drawing information; adjust the size of the three-dimensional sub-object according to the joint location information of the three-dimensional sub-object, and obtain an adjustment The three-dimensional sub-object after the combination; the combined three-dimensional sub-object obtains the initial three-dimensional object; and the initial three-dimensional object is adjusted according to the minimized objective function to obtain a three-dimensional object. A three-dimensional object reconstruction terminal 130 according to an embodiment of the present invention is described below. The terminal 130 includes: a three-dimensional object reconstruction device 131, a display 132, wherein the display 132 is used to display a three-dimensional object. A person skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the terminal, the device and the unit described above can refer to the corresponding process in the foregoing method embodiment, and details are not described herein again. In the several embodiments provided in the present application, it should be understood that the disclosed terminal, apparatus, and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form. In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product. Based on such understanding, the technical solution of the present invention, which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including The instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read only memory (ROM, ead-Only Memory), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program code. .

以上所述，仅为本发明较佳的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到的变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应该以权利要求的保护范围为准下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。 The above description is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or within the technical scope disclosed by the present invention. Alternatives are intended to be covered by the scope of the present invention. Therefore, the scope of the present invention should be clearly understood that the technical solutions in the embodiments of the present invention are clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It is only a part of the embodiments of the invention, not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

Claims

权利要求 Rights request

1、三维物体的二维线画图信息，所述二维线画图信息包括二维线画图；根据所述二维线画图信息，按照自由度分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数；根据所述线画图子图重建三维子物体；组合全部的所述三维子物体得到三维物体。 1. Two-dimensional line drawing information of a three-dimensional object. The two-dimensional line drawing information includes a two-dimensional line drawing. According to the two-dimensional line drawing information, the two-dimensional line drawing is divided according to degrees of freedom to obtain at least one line drawing. Figure, where the degrees of freedom are the minimum number of vertices that need to be known to determine the spatial position of a three-dimensional object composed of planes; reconstruct the three-dimensional sub-object according to the line drawing sub-image; combine all the three-dimensional sub-objects Objects get three-dimensional objects.

2、如权利要求 1所述的三维物体重建的方法，其特征在于，每一个所述线画图子图对应一个三维子物体；在全部所述线画图子图中，任意一个线画图子图所表示的三维子物体的最小的自由度为 4。 2. The method for reconstructing a three-dimensional object according to claim 1, wherein each of the line drawing sub-pictures corresponds to a three-dimensional sub-object; in all of the line drawing sub-pictures, any one of the line drawing sub-pictures is The minimum degree of freedom of the three-dimensional sub-object represented is 4.

3、如权利要求 1或 2所述的三维物体重建的方法，其特征在于，所述根据所述二维线画图信息，按照自由度分割所述二维线画图得到至少一个线画图子图包括：根据所述二维线画图信息得到二维线画图的顶点坐标、顶点之间的线段信息，根据所述二维线画图的顶点坐标和所述顶点之间的线段信息得到二维线画图的面信息；根据所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图。 3. The three-dimensional object reconstruction method according to claim 1 or 2, characterized in that, according to the two-dimensional line drawing information, dividing the two-dimensional line drawing according to degrees of freedom to obtain at least one line drawing sub-image includes : Obtain the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices according to the two-dimensional line drawing information, and obtain the two-dimensional line drawing based on the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices. Surface information; According to the surface information of the two-dimensional line drawing, divide the two-dimensional line drawing according to the degree of freedom to obtain at least one sub-picture of the line drawing.

4、如权利要求 3所述的三维物体重建的方法，其特征在于，所述二维线画图的面信息包括二维线画图的面；所述根据所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图，具体包括：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面； 4. The method for reconstructing a three-dimensional object according to claim 3, wherein the surface information of the two-dimensional line drawing includes the surface of the two-dimensional line drawing; the surface information based on the two-dimensional line drawing is according to The degree of freedom divides the two-dimensional line drawing to obtain at least one sub-image of the line drawing, which specifically includes: selecting any pair of adjacent faces that have not been selected among all the faces of the two-dimensional line drawing. , as the first line drawing surface and the second line drawing surface; Set the first set as an empty set, calibrate the first line drawing surface and the second line drawing surface as determined surfaces and add them to the first set; among all the two-dimensional line drawing surfaces, Other line drawing surfaces except the first line drawing surface and the second line drawing surface are calibrated as uncertain surfaces;

在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图； Select a third line drawing surface from the uncertain surface, calibrate the third line drawing surface as a determined surface and add it to the first set, repeat this step until all the uncertain surfaces are traversed, where , the third line drawing surface has at least three non-collinear vertices, and the non-collinear vertices are the vertices of the line drawing surface in the first set; one is obtained according to the determined surface in the first set Corresponding line drawing sub-picture;

重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。 Repeat the above steps until all adjacent faces in all the faces of the two-dimensional line drawing have been selected once.

5、如权利要求 4所述的三维物体重建的方法，其特征在于，如果所述第一集合中的确定面的数目不小于三，在根据所述第一集合中的确定面得到一个对应的线画图子图之后，所述方法还包括：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。 5. The three-dimensional object reconstruction method according to claim 4, characterized in that, if the number of determined surfaces in the first set is not less than three, a corresponding one is obtained according to the determined surfaces in the first set. After the line drawing sub-pictures, the method further includes: adding the line drawing sub-pictures represented by the first set to the second set; obtaining a coexistence relationship graph based on all the line drawing sub-pictures in the second set, Wherein, one vertex of the coexistence relationship graph represents one of the line drawing subgraphs, and the number of faces of the line drawing subgraph is used as the weight of the vertex. If two of the line drawing subgraphs If there is a coexistence relationship between them, then there is a connection between the vertices of the two corresponding line drawing subgraphs; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing subgraph, and the coexistence relationship The relationship means that the objects represented by the two line drawing sub-figures do not share any plane or the common plane between the objects represented by the two line drawing sub-figures only exists on the same plane; Find the maximum weight cluster in the above-mentioned coexistence relationship graph, and determine the line drawing subgraph corresponding to the vertex of the maximum weight cluster, where clique is a graph with vertices and edges, and between any two vertices in the cluster are all connected by an edge, and the maximum weight group is the sum of weights in all groups of the coexistence relationship graph. The largest group.

6、如权利要求 1-5任意之一所述的三维物体重建的方法，其特征在于，所述根据所述线画图子图重建三维子物体，包括： 6. The method of reconstructing a three-dimensional object according to any one of claims 1 to 5, characterized in that the reconstruction of a three-dimensional sub-object based on the line drawing sub-image includes:

根据最小化目标函数得到所述线画图子图的顶点的深度值； Obtain the depth value of the vertex of the line drawing subgraph according to the minimization objective function;

获取所述线画图子图的顶点坐标； Obtain the vertex coordinates of the line drawing subgraph;

获取三维子物体的面； Get the surface of the three-dimensional sub-object;

根据所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。 The three-dimensional sub-object is obtained according to the depth value of the vertex of the line drawing sub-graph, the vertex coordinate of the line drawing sub-graph and the surface of the three-dimensional sub-object.

7、如权利要求 1-6任意之一所述的三维物体重建的方法，其特征在于，所述组合全部的所述三维子物体得到所述三维物体，包括：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式： 7. The three-dimensional object reconstruction method according to any one of claims 1 to 6, characterized in that said combining all the three-dimensional sub-objects to obtain the three-dimensional object includes: combining all the three-dimensional sub-objects by combining parts. The three-dimensional sub-object is obtained by obtaining the three-dimensional object, and the method of combining the parts includes at least one of the following methods:

点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。 Point-to-point combination, point-to-line combination, point-to-surface combination, line-to-line combination, line-to-surface combination, surface-to-surface combination, and point-to-line combination.

8、如权利要求 7所述的三维物体重建的方法，其特征在于，所述通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，具体包括：根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体； 8. The method for reconstructing a three-dimensional object according to claim 7, characterized in that: combining all the three-dimensional sub-objects by combining parts to obtain the three-dimensional object specifically includes: according to the two-dimensional line The drawing information obtains the combination part information of the three-dimensional sub-object; adjusts the size of the three-dimensional sub-object according to the combination part information of the three-dimensional sub-object, and obtains the adjusted three-dimensional sub-object;

组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。 Combining the adjusted three-dimensional sub-objects to obtain an initial three-dimensional object; adjusting the initial three-dimensional object according to the minimization objective function to obtain the three-dimensional object.

9、如权利要求 1-8任意之一所述的三维物体重建的方法，其特征在于，通过下列之一的方式获取所述三维物体的二维线画图信息： 9. The three-dimensional object reconstruction method according to any one of claims 1 to 8, characterized in that the two-dimensional line drawing information of the three-dimensional object is obtained in one of the following ways:

通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式和通过从预先存储的照片中提取线画图的方式。 Shooting with a camera, input with a touch device, input with a mouse formula and drawing by extracting lines from pre-stored photos.

10、一种三维物体重建的方法，其特征在于，所述方法包括：获取三维物体的二维线画图，以及所述二维线画图中的顶点坐标、顶点之间的线段信息；根据所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图；根据每一个所述线画图子图重建一个对应的三维子物体；组合全部的所述三维子物体得到三维物体。 10. A method for reconstructing a three-dimensional object, characterized in that the method includes: obtaining a two-dimensional line drawing of the three-dimensional object, and the vertex coordinates and line segment information between the vertices in the two-dimensional line drawing; according to the Segment the two-dimensional line drawing by the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices to obtain at least one line drawing sub-graph; reconstruct a corresponding three-dimensional sub-object based on each of the line drawing sub-graphs; A three-dimensional object is obtained by combining all the three-dimensional sub-objects.

11、如权利要求 10所述的三维物体重建的方法，其特征在于，所述根据所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图，具体包括：根据所述二维线画图的顶点和所述顶点之间的线段信息获得二维线画图的面信息；根据所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图。 11. The method for reconstructing a three-dimensional object according to claim 10, wherein the two-dimensional line drawing is divided according to the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices to obtain at least A line drawing sub-graph specifically includes: obtaining the surface information of the two-dimensional line drawing according to the vertices of the two-dimensional line drawing and the line segment information between the vertices; segmenting the two-dimensional line drawing according to the surface information of the two-dimensional line drawing A two-dimensional line drawing is performed to obtain at least one subgraph of the line drawing.

12、如权利要求 11所述的三维物体重建的方法，其特征在于，所述根据所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图包括：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。 12. The method for reconstructing a three-dimensional object according to claim 11, wherein the two-dimensional line drawing is segmented according to the surface information of the two-dimensional line drawing to obtain at least one sub-image of the line drawing including : Select any pair of adjacent surfaces that have not been selected among all the two-dimensional line drawing surfaces as the first line drawing surface and the second line drawing surface; Set the first set as an empty set, and set the The first line drawing surface and the second line drawing surface are calibrated as certain surfaces and added to the first set; among all the two-dimensional line drawing surfaces, except the first line drawing surface and the The other line drawing surfaces of the second line drawing surface are calibrated as uncertainty surfaces; Select a third line drawing surface from the uncertain surface, calibrate the third line drawing surface as a determined surface and add it to the first set, repeat this step until all the uncertain surfaces are traversed, where , the third line drawing surface has at least three non-collinear vertices, and the non-collinear vertices are the vertices of the line drawing surface in the first set; one is obtained according to the determined surface in the first set Corresponding line drawing sub-picture; Repeat the above steps until all adjacent faces in all the two-dimensional line drawing faces have been selected once.

13、如权利要求 12所述的方法，其特征在于，如果所述第一集合中的确定面的数目不小于三，在根据所述第一集合中的确定面得到一个对应的线画图子图之后，所述方法还包括：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。 13. The method of claim 12, wherein if the number of certain surfaces in the first set is not less than three, a corresponding line drawing sub-picture is obtained according to the certain surfaces in the first set. After that, the method further includes: adding the line drawing sub-pictures represented by the first set to the second set; obtaining the coexistence relationship graph according to all the line drawing sub-pictures in the second set, wherein: A vertex of the coexistence relationship graph represents one of the line drawing subgraphs, and the number of faces of the line drawing subgraph is used as the weight of the vertex. If there is coexistence between two of the line drawing subgraphs relationship, then there is a connection between the vertices of the two corresponding line drawing subgraphs; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing subgraph, and the coexistence relationship refers to the The objects represented by the two line drawing sub-figures do not share any plane or the common plane between the objects represented by the two line drawing sub-figures only exists on the same plane; in the coexistence relationship diagram Find the maximum weight group in , and determine the line drawing subgraph corresponding to the vertex of the maximum weight group. Where, the group is a graph with vertices and edges, and there is an edge between any two vertices in the group. Connection, the maximum weight group is the group with the largest sum of weights among all groups in the coexistence relationship graph.

14、如权利要求 10-13任意之一所述的三维物体重建的方法，其特征在于，所述根据一个所述线画图子图重建一个对应的三维子物体，包括：根据最小化目标函数得到所述线画图子图的顶点的深度值；获取所述线画图子图的顶点坐标；获取三维子物体的面；根据所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。 14. The three-dimensional object reconstruction method according to any one of claims 10 to 13, characterized in that: reconstructing a corresponding three-dimensional sub-object according to one of the line drawing sub-pictures includes: obtaining according to the minimization objective function The depth value of the vertex of the line drawing subgraph; Obtain the vertex coordinates of the line drawing sub-graph; Obtain the surface of the three-dimensional sub-object; Obtain the obtained result based on the depth value of the vertex of the line drawing sub-graph, the vertex coordinates of the line drawing sub-graph and the surface of the three-dimensional sub-object Describe the three-dimensional sub-object.

15、如权利要求 10-14任意之一所述的三维物体重建的方法，其特征在于，所述组合全部的所述三维子物体得到所述三维物体，包括：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。 15. The three-dimensional object reconstruction method according to any one of claims 10 to 14, wherein said combining all the three-dimensional sub-objects to obtain the three-dimensional object includes: combining all the three-dimensional sub-objects by combining parts. The three-dimensional sub-object is obtained by obtaining the three-dimensional object, and the way of combining the parts includes at least one of the following ways: point-point combination, point-line combination, point-surface combination, line-line combination, line-surface combination, surface-surface combination and point-to-point combination Line and surface combination.

16、如权利要求 15所述的三维物体重建的方法，其特征在于，所述通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，具体包括：根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。 16. The method for reconstructing a three-dimensional object according to claim 15, characterized in that: combining all the three-dimensional sub-objects by combining parts to obtain the three-dimensional object specifically includes: according to the two-dimensional line The drawing information is used to obtain the combination part information of the three-dimensional sub-object; the size of the three-dimensional sub-object is adjusted according to the combination part information of the three-dimensional sub-object to obtain the adjusted three-dimensional sub-object; and the adjusted three-dimensional sub-object is obtained by combining the Initial three-dimensional object; Adjust the initial three-dimensional object according to the minimization objective function to obtain the three-dimensional object.

17、如权利要求 10-16任意之一所述的三维物体重建的方法，其特征在于，通过下列之一的方式获取所述二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式、通过从预先存储的照片中提取线画图的方式。 17. The three-dimensional object reconstruction method according to any one of claims 10 to 16, characterized in that the two-dimensional line drawing information is obtained in one of the following ways: by shooting with a camera device, or by inputting with a touch device method, by mouse input, and by extracting line drawings from pre-stored photos.

18、一种三维物体重建装置，其特征在于，所述装置包括：获取模块，用于获取三维物体的二维线画图信息，所述二维线画图信息包括二维线画图；分割模块，用于根据所述获取模块获取的所述二维线画图信息，按照自由度分割所述二维线画图，得到至少一个线画图子图，其中，所述自由度为确定一个由平面组成的三维物体的空间位置所需要知道的最少的顶点的个数；重建模块，用于根据所述分割模块得到的所述线画图子图重建三维子物体；组合模块，用于组合全部的所述重建模块得到的所述三维子物体得到三维 18. A three-dimensional object reconstruction device, characterized in that the device includes: an acquisition module for acquiring two-dimensional line drawing information of a three-dimensional object, where the two-dimensional line drawing information includes a two-dimensional line drawing; A segmentation module, configured to segment the two-dimensional line drawing according to degrees of freedom according to the two-dimensional line drawing information obtained by the acquisition module, to obtain at least one line drawing sub-graph, wherein the degree of freedom is determined by a plane The minimum number of vertices that need to be known for the spatial position of the composed three-dimensional object; a reconstruction module, used to reconstruct the three-dimensional sub-object based on the line drawing sub-image obtained by the segmentation module; a combination module, used to combine all the The three-dimensional sub-object obtained by the reconstruction module obtains a three-dimensional

19、如权利要求 18所述的三维物体重建装置，其特征在于，所述分割模块包括：面信息获取单元，用于根据所述获取模块获得的所述二维线画图信息得到二维线画图的顶点坐标、顶点之间的线段信息，根据所述二维线画图的顶点坐标和所述顶点之间的线段信息得到二维线画图的面信息；面分割单元，用于根据所述面信息获取单元得到的所述二维线画图的面信息，按照所述自由度分割所述二维线画图，得到至少一个所述线画图子图。 19. The three-dimensional object reconstruction device according to claim 18, wherein the segmentation module includes: a surface information acquisition unit, configured to obtain a two-dimensional line drawing based on the two-dimensional line drawing information obtained by the acquisition module. The vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices are used to obtain the surface information of the two-dimensional line drawing according to the vertex coordinates of the two-dimensional line drawing and the line segment information between the vertices; A surface segmentation unit is used to obtain the surface information of the two-dimensional line drawing according to the surface information The acquisition unit obtains the surface information of the two-dimensional line drawing, divides the two-dimensional line drawing according to the degree of freedom, and obtains at least one sub-image of the line drawing.

20、如权利要求 19所述的三维物体重建装置，其特征在于，所述面分割单元具体用于：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。 20. The three-dimensional object reconstruction device according to claim 19, characterized in that the surface segmentation unit is specifically used to: select any pair of adjacent adjacent ones that have not been selected among all the surfaces of the two-dimensional line drawing. as the first line drawing surface and the second line drawing surface; set the first set as an empty set, calibrate the first line drawing surface and the second line drawing surface as determined surfaces and add them to the first set in; among all the two-dimensional line drawing surfaces, calibrate the other line drawing surfaces except the first line drawing surface and the second line drawing surface as uncertain surfaces; select the third line drawing surface among the uncertain surfaces; For a three-line drawing surface, calibrate the third line drawing surface as a determined surface and add it to the first set. Repeat this step until all the uncertain surfaces have been traversed, where the third line drawing surface has at least three non-collinear vertices, and the non-collinear vertices are the first The vertices of the line drawing surfaces in the set; Obtain a corresponding line drawing sub-picture according to the determined surface in the first set; Repeat the above steps until all adjacent surfaces in all the two-dimensional line drawing surfaces are Each pair has been selected once.

21、如权利要求 20所述的三维物体重建装置，其特征在于，如果所述第一集合中的确定面的数目不小于三，所述面分割单元还用于：将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。 21. The three-dimensional object reconstruction device according to claim 20, wherein if the number of determined faces in the first set is not less than three, the face segmentation unit is further configured to: The line drawing subgraph represented is added to the second set; a coexistence relationship graph is obtained according to all the line drawing subgraphs in the second set, wherein one vertex of the coexistence relationship graph represents one of the line drawing subgraphs, The number of faces of the line drawing subgraph is used as the weight of the vertex. If there is a coexistence relationship between the two line drawing subgraphs, then corresponding to one of the vertices of the two line drawing subgraphs There are connections between them; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing sub-pictures, and the coexistence relationship means that there is no sharing between the objects represented by the two line drawing sub-pictures. The common plane between any plane or the objects represented by the two line drawing sub-graphs only exists on the same plane; find the maximum weight group in the coexistence relationship diagram and determine the maximum weight group The line drawing subgraph corresponding to the vertex, where clique is a graph with vertices and edges, any two vertices in the clique are connected by an edge, and the maximum weight clique is in the coexistence relationship graph The group with the largest sum of weights among all groups.

22、如权利要求 18-21任意之一所述的三维物体重建的装置，其特征在于，所述重建模块包括：重建信息获取单元，用于根据最小化目标函数得到所述线画图子图的顶点的深度值，获取所述线画图子图的顶点坐标，获取三维子物体的面；重建单元，用于根据所述重建信息获取单元得到的所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。 22. The three-dimensional object reconstruction device according to any one of claims 18 to 21, characterized in that the reconstruction module includes: a reconstruction information acquisition unit, used to obtain the line drawing sub-image according to the minimization objective function. The depth value of the vertex, obtains the vertex coordinates of the line drawing sub-graph, and obtains the surface of the three-dimensional sub-object; The reconstruction unit is used to obtain the depth value of the vertex of the line drawing sub-graph and the depth value of the vertex of the line drawing sub-graph obtained by the reconstruction information acquisition unit. The three-dimensional sub-object is obtained by combining the vertex coordinates of the line drawing sub-graph and the surface of the three-dimensional sub-object.

23、如权利要求 18-22任意之一所述的三维物体重建的装置，其特征在于，所述组合模块具体用于：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式： 23. The device for reconstructing a three-dimensional object according to any one of claims 18 to 22, characterized in that the combination module is specifically used to: combine all the three-dimensional sub-objects by combining parts to obtain the three-dimensional object. Object, the method of combining the parts includes at least one of the following methods:

24、如权利要求 18-22任意之一所述的三维物体重建的装置，其特征在于，所述组合模块具体用于：根据所述二维线画图信息得到所述三维子物体的结合部位信息； 24. The device for reconstructing a three-dimensional object according to any one of claims 18 to 22, characterized in that the combination module is specifically used to: obtain the combination part information of the three-dimensional sub-object according to the two-dimensional line drawing information. ;

根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体； Adjust the size of the three-dimensional sub-object according to the combination part information of the three-dimensional sub-object to obtain the adjusted three-dimensional sub-object;

25、如权利要求 18-24任意之一所述的三维物体重建的装置，其特征在于，所述获取模块通过下列之一的方式获取所述三维物体的二维线画图信息：通过摄像设备拍摄的方式、通过触摸设备输入的方式、通过鼠标输入的方式和通过从预先存储的照片中提取线画图的方式。 25. The device for reconstructing a three-dimensional object according to any one of claims 18 to 24, characterized in that the acquisition module obtains the two-dimensional line drawing information of the three-dimensional object in one of the following ways: shooting with a camera device , input via a touch device, input via a mouse, and drawing by extracting line drawings from pre-stored photos.

26、一种三维物体重建的装置，其特征在于，所述装置包括：信息获取模块，用于获取三维物体的二维线画图，以及所述二维线画图中的顶点坐标、顶点之间的线段信息；线画图子图分割模块，用于根据所述信息获取模块获取的所述二维线画图的顶点坐标和所述顶点之间的线段信息分割所述二维线画图，得到至少一个的线画图子图；三维子物体重建模块，用于根据所述线画图子图分割模块得到的每一个所述线画图子图重建一个对应的三维子物体；三维物体组合模块，用于组合所述三维子物体重建模块重建的全部的所述三维子物体得到三维物体。 26. A device for reconstructing a three-dimensional object, characterized in that the device includes: an information acquisition module for acquiring a two-dimensional line drawing of a three-dimensional object, as well as the vertex coordinates and the distance between the vertices in the two-dimensional line drawing. Line segment information; a line drawing sub-graph segmentation module, configured to segment the two-dimensional line drawing according to the vertex coordinates of the two-dimensional line drawing obtained by the information acquisition module and the line segment information between the vertices, and obtain at least one Line drawing sub-picture; a three-dimensional sub-object reconstruction module, used for each of the sub-pictures obtained according to the line drawing sub-picture segmentation module The line drawing sub-picture reconstructs a corresponding three-dimensional sub-object; a three-dimensional object combination module is used to combine all the three-dimensional sub-objects reconstructed by the three-dimensional sub-object reconstruction module to obtain a three-dimensional object.

27、如权利要求 26所述的三维物体重建的装置，其特征在于，所述线画图子图分割模块，包括：二维线画图的面信息获取单元，根据所述信息获取模块获取的二维线画图的顶点和所述顶点之间的线段信息获得二维线画图的面信息；线画图子图分割单元，用于根据所述二维线画图的面信息获取单元获取的所述二维线画图的面信息分割所述二维线画图，得到至少一个的所述线画图子图。 27. The device for three-dimensional object reconstruction according to claim 26, characterized in that the line drawing sub-image segmentation module includes: a two-dimensional line drawing surface information acquisition unit, based on the two-dimensional line drawing obtained by the information acquisition module. The vertex of the line drawing and the line segment information between the vertices obtain the surface information of the two-dimensional line drawing; the line drawing sub-picture dividing unit is used to obtain the two-dimensional line according to the surface information acquisition unit of the two-dimensional line drawing The surface information of the drawing is used to divide the two-dimensional line drawing to obtain at least one sub-graph of the line drawing.

28、如权利要求 27所述的三维物体重建的装置，其特征在于，所述线画图子图分割单元具体用于：在所述全部的二维线画图的面中选择任意一对还没有选择过的相邻的面，作为第一线画图面和第二线画图面；设置第一集合为空集，将所述第一线画图面和所述第二线画图面标定为确定面并加入到所述第一集合中；在全部的所述二维线画图的面中将除所述第一线画图面和所述第二线画图面的其它线画图面标定为不确定面；在所述不确定面中选择第三线画图面，将所述第三线画图面标定为确定面并加入到所述第一集合中，重复本歩骤直到遍历完全部的所述不确定面，其中，所述第三线画图面至少有三个不共线的顶点，且所述不共线的顶点是所述第一集合中的线画图面的顶点；根据所述第一集合中的确定面得到一个对应的线画图子图；重复以上歩骤，直到所述全部的二维线画图的面中的所有相邻面对都被选择过一次。 28. The device for three-dimensional object reconstruction according to claim 27, wherein the line drawing sub-image segmentation unit is specifically used to: select any pair that has not yet been selected among all the two-dimensional line drawing faces. The adjacent surfaces passed through are used as the first line drawing surface and the second line drawing surface; set the first set as an empty set, calibrate the first line drawing surface and the second line drawing surface as determined surfaces and add them to all In the first set; among all the two-dimensional line drawing surfaces, other line drawing surfaces except the first line drawing surface and the second line drawing surface are calibrated as uncertain surfaces; in the uncertain surfaces Select the third line drawing surface from the surface, calibrate the third line drawing surface as a determined surface and add it to the first set, repeat this step until all the uncertain surfaces are traversed, where, the third line drawing surface The drawing surface has at least three non-collinear vertices, and the non-collinear vertices are the vertices of the line drawing surface in the first set; a corresponding line drawing is obtained according to the determined surface in the first set Figure; Repeat the above steps until all adjacent faces in all the faces of the two-dimensional line drawing have been selected once.

29、如权利要求 28所述的装置，其特征在于，所述线画图子图分割单元还用于：如果所述第一集合中的确定面的数目不小于三，则将所述第一集合所表示的线画图子图加入到第二集合中；根据所述第二集合中所有的线画图子图得到共存关系图，其中，所述共存关系图的一个顶点表示一个所述线画图子图，所述线画图子图所具有的面的个数做为所述顶点的权值，如果两个所述线画图子图之间存在共存关系，则对应两个所述线画图子图的顶点之间有连线连接；所述共存关系图表示所述线画图子图所表示的物体之间的共存关系，所述共存关系是指所述两个线画图子图所表示的物体之间没有共享任何一个平面或者所述两个线画图子图所表示的物体之间的公共面只存在于同一个平面上；在所述共存关系图中找到最大权值团，确定所述最大权值团的顶点对应的线画图子图，其中，团是一个拥有顶点和边的图，所述团中的任意两个顶点之间都有一条边连接，所述最大权值团是在所述共存关系图的所有团中权值之和最大的团。 29. The device of claim 28, wherein the line drawing sub-picture dividing unit is further configured to: if the number of determined faces in the first set is not less than three, divide the first set into The represented line drawing subgraph is added to the second set; a coexistence relationship graph is obtained according to all the line drawing subgraphs in the second set, wherein one vertex of the coexistence relationship graph represents one of the line drawing subgraphs , the number of faces of the line drawing subgraph is used as the weight of the vertex. If there is a coexistence relationship between the two line drawing subgraphs, then the corresponding vertices of the two line drawing subgraphs There is a connection between them; the coexistence relationship diagram represents the coexistence relationship between the objects represented by the line drawing sub-pictures, and the coexistence relationship means that there is no coexistence relationship between the objects represented by the two line drawing sub-pictures. Sharing any plane or the common plane between the objects represented by the two line drawing sub-graphs only exists on the same plane; Find the maximum weight group in the coexistence relationship diagram and determine the maximum weight group The line drawing subgraph corresponding to the vertex of The group with the largest sum of weights among all groups in the graph.

30、如权利要求 26-29任意之一所述的三维物体重建的装置，其特征在于，所述三维子物体重建模块包括：三维子物体重建信息获取单元，用于根据最小化目标函数得到所述线画图子图的顶点的深度值，获取所述线画图子图的顶点坐标，获取三维子物体的面；三维子物体重建单元，用于根据所述三维子物体重建信息获取单元得到的所述线画图子图的顶点的深度值、所述线画图子图的顶点坐标和所述三维子物体的面得到所述三维子物体。 30. The three-dimensional object reconstruction device according to any one of claims 26 to 29, characterized in that the three-dimensional sub-object reconstruction module includes: a three-dimensional sub-object reconstruction information acquisition unit, used to obtain the obtained object according to the minimization objective function. The depth value of the vertex of the line drawing sub-graph is obtained, the vertex coordinates of the line drawing sub-graph are obtained, and the surface of the three-dimensional sub-object is obtained; A three-dimensional sub-object reconstruction unit is used to obtain all the data obtained by the three-dimensional sub-object reconstruction information acquisition unit. The depth value of the vertex of the line drawing subgraph, the vertex coordinate of the line drawing subgraph and the surface of the three-dimensional sub-object are used to obtain the three-dimensional sub-object.

31、如权利要求 26-30任意之一所述的三维物体重建的装置，其特征在于，所述三维物体组合模块具体用于：通过部位结合的方式组合全部的所述三维子物体，得到所述三维物体，所述部位结合的方式包括下列至少之一的方式：点点结合、点线结合、点面结合、线线结合、线面结合、面面结合和点线面结合。 31. The three-dimensional object reconstruction device according to any one of claims 26 to 30, characterized in that the three-dimensional object combination module is specifically used to: combine all the three-dimensional sub-objects by combining parts to obtain the obtained Describing a three-dimensional object, the The method of combining the parts includes at least one of the following methods: point-point combination, point-line combination, point-surface combination, line-line combination, line-surface combination, surface-surface combination and point-line-surface combination.

32、如权利要求 26-30任意之一所述的三维物体重建的装置，其特征在于，所述三维物体组合模块具体用于：根据所述二维线画图信息得到所述三维子物体的结合部位信息；根据所述三维子物体的结合部位信息调整所述三维子物体的大小，得到调整后的三维子物体；组合所述调整后的三维子物体得到初始三维物体；根据最小化目标函数调整所述初始三维物体，得到所述三维物体。 32. The device for three-dimensional object reconstruction according to any one of claims 26 to 30, characterized in that the three-dimensional object combination module is specifically used to: obtain the combination of the three-dimensional sub-objects based on the two-dimensional line drawing information. Part information; Adjust the size of the three-dimensional sub-object according to the combined part information of the three-dimensional sub-object to obtain an adjusted three-dimensional sub-object; Combine the adjusted three-dimensional sub-objects to obtain an initial three-dimensional object; Adjust according to the minimization objective function The initial three-dimensional object is used to obtain the three-dimensional object.

33、一种三维物体重建的终端，其特征在于，所述终端包括：如权利要求 18-32任意之一所述的三维物体重建装置，显示器，其中，所述显示器用于显示所述三维物体。 33. A terminal for three-dimensional object reconstruction, characterized in that the terminal includes: the three-dimensional object reconstruction device according to any one of claims 18 to 32, a display, wherein the display is used to display the three-dimensional object .